universitÉ du quÉbec À rimouski - uqar
TRANSCRIPT
UNIVERSITÉ DU QUÉBEC À RIMOUSKI
MÉCANISMES DE SÉQUESTRATION ET CINÉTIQUE DE
BIOACCUMULATION ET DE MÉTABOLISATION DES HYDROCARBURES
AROMATIQUES POLYCYCLIQUES (HAP) DANS LES SÉDIMENTS MARINS ET
LACUSTRES
THÈSE
PRÉSENTÉ À
L'UNIVERSITÉ DU QUÉBEC À RIMOUSKI
Comme exigence partielle
du programme de doctorat en océanographie
PHILOSOPHIAE DOCTOR (OCÉANOGRAPHIE)
PAR
MICKAËL BARTHE
Mai 2008
UNIVERSITÉ DU QUÉBEC À RIMOUSKI Service de la bibliothèque
Avertissement
La diffusion de ce mémoire ou de cette thèse se fait dans le respect des droits de son auteur, qui a signé le formulaire « Autorisation de reproduire et de diffuser un rapport, un mémoire ou une thèse ». En signant ce formulaire, l’auteur concède à l’Université du Québec à Rimouski une licence non exclusive d’utilisation et de publication de la totalité ou d’une partie importante de son travail de recherche pour des fins pédagogiques et non commerciales. Plus précisément, l’auteur autorise l’Université du Québec à Rimouski à reproduire, diffuser, prêter, distribuer ou vendre des copies de son travail de recherche à des fins non commerciales sur quelque support que ce soit, y compris l’Internet. Cette licence et cette autorisation n’entraînent pas une renonciation de la part de l’auteur à ses droits moraux ni à ses droits de propriété intellectuelle. Sauf entente contraire, l’auteur conserve la liberté de diffuser et de commercialiser ou non ce travail dont il possède un exemplaire.
REMERCIEMENTS
De nombreuses personnes m'ont aidé et soutenu pendant ces années de doctorat.
l'aimerais remercier plus particulièrement:
Mon directeur de thèse, le docteur Émilien Pelletier, pour sa totale confiance, ses
encouragements stimulants, son accompagnement dans mon processus de réflexion, sa
compétence et son soutien financier.
Les Dr Gijs D. Breedveld et Gerard Cornelissen, chercheur au Norwegian
Geotechnical Institute pour m'avoir accueilli et aidé lors de mon séjour à Oslo.
Le Dr Claude Rouleau, chercheur à l'Institut Maurice Lamontagne, pour m ' avoir aidé
pour les cinétiques des métabolites.
Les alumineries Alcan Ltée pour avoir fournit les échantillons du Fjord de Kitimat et
de la Rivière Saint-Louis.
Un immense merci à ma chère et tendre Zazou pour son support, ses encouragements
et surtout pour sa patience.
Il
RÉSUMÉ
Traditionnellement, les concentrations en hydrocarbures aromatiques polycycliques
CHAP) dans les sols et les sédiments ont été déterminées après des extractions solide/liquide
vigoureuses utilisant des solvants organiques, tels que le dichlorométhane ou
l'hexane/acétone. Cependant, quand les contaminants organiques hydrophobes, tels que les
HAP, entrent dans les sols ou sédiments, ils subissent un certain nombre de processus de
perte ou de transport qui peuvent modifier leur comportement. De plus, les processus de
sorption peuvent permettre à ces composés chimiques de devenir plus résistants à
l'extraction par des solvants. Les résultats de ce processus d'altération avec le temps se
traduisent par une diminution de l'extractabilité des HAP et un déclin de leur disponibilité
pour l' ingestion par les organismes benthiques et leur biodégradabilité par les
microorganismes. Ce phénomène a été appelé "effet de vieillissement". Un effort important
en recherche a été récemment consacré au développement de méthodes chimiques fiables
afin de déterminer la partie biodisponible présente dans les sols et sédiments.
L'objectif de ce travail a donc été de déterminer une ou plusieurs méthodes chimiques
dites sélectives permettant de caractériser les concentrations en HAP biodisponibles dans
des sédiments dont le niveau de contamination et de propriétés géochimiques étaient
variables. De plus, nous avons travaillé à établir les patrons temporels d' accumulation des
HAP de ces échantillons dans deux espèces de vers en identifiant les paramètres
toxicocinétiques décrivant l'ingestion, l'élimination et les facteurs de bioaccumulation.
111
Ainsi, trois séries d'expériences ont été effectuées. La première a permis de comparer
trois différentes méthodes d'extraction solide/liquide (n-butanol (BuOH), hydroxypropyl-~
cyclodextrine (HPCD), et une solution du tensioactif Brij700 (B700)) et la seconde de
comparer l'efficacité de deux échantillonneurs passifs, des bandes de polyoxyméthylène
(POM) et des tubes en silicone de polydimethylsiloxane (PDMS). Les résultats de ces deux
études ont été comparés avec les résultats de tests de bioaccumulation sur 28 jours effectués
sur les mêmes sédiments avec Nereis virens pour les sédiments marins et Lumbriculus
variegatus pour les sédiments lacustres. De plus, les résultats de la seconde étude ont été
comparés avec ceux obtenus lors de la première étude avec la solution aqueuse de B700. La
troisième expérience a permis de déterminer les patrons temporels d'accumulation des
différents HAP par les vers (N. virens et L. variegatus) dans les différents sédiments étudiés
ainsi que l 'habileté des organismes à métaboliser le phénanthrène et le pyrène par la
détermination des hydroxy-HAP correspondants (9-hydroxyphénanthrène et 1-
hydroxypyrène ).
Les résultats obtenus lors des deux premières études montrent que des différences
importantes ont été observées aussi bien dans la quantité que dans les proportions des HAP
obtenues en utilisant différentes techniques pour déterminer la biodisponibilité des HAP
dans des sédiments. Notre première étude montre que le BuOH extrayait plus que ce que les
vers pouvaient bioaccumuler dans leurs tissus, et que le HPCD sousestimait la
biodisponibilité des HAP spécialement les HAP de faible poids moléculaire. D'autre part,
l' utilisation du tensioactif B700 a fourni une bonne prédiction pour les HAP dans les
sédiments fortement contaminés. Les résultats de la seconde étude montrent que le PDMS
IV
surestime la disponibilité des HAP dans les sédiments étudiés (faiblement et fortement
contaminés), alors que le POM produit des résultats similaires à l'extraction au B700. En
effet, la biodisponibilité des HAP totaux a été prédite correctement par le POM et le B700
dans les sédiments fortement contaminés par des alumineries. Par contre, un examen plus
détaillé des résultats obtenus pour les HAP individuels indique que les deux techniques
surestiment la disponibilité des grosses molécules avec un log Kow >6 suggérant un
mécanisme biologique limitant l' incorporation des plus grosses molécules de HAP ce qui
semble être relié à la taille moléculaire des composés ou à leur encombrement stérique.
Les résultats de la troisième étude ont montré que Nereis virens et Lumbriculus
variegatus accumulaient rapidement les HAP associés au sédiment et atteignaient un état-
stable apparent en sept jours (168 h) pour tous les sédiments étudiés. De plus, les constantes
du taux d'ingestion des HAP sont plus faibles pour les sédiments fortement contaminés que
pour les faiblement contaminés (10-2_10-6 et 10-1_10-4 g sédiment sec g-I organisme humide
h- I, respectivement), alors que les constantes du taux d'élimination des HAP sont du même
ordre de grandeur pour les deux types de sédiments. De plus, celle-ci a montré que le log de
la constante du taux d ' ingestion du HAP par l' organisme (log ks) diminuée avec
l' augmentation de l'hydrophobicité des HAP, exprimée par le coefficient de partition
octanol-eau log Kow pour tous les sédiments étudiés. La relation négative entre log ks et log
Kow suggère que le taux de désorption du composé à partir du sédiment était un facteur
important dans l'accumulation par les vers. La détermination de la proportion relative des
métabolites de phase 1 par rapport au phénanthrène total et au pyrène total dans les tissus de
vers après une exposition de 28 jours indiquait la présence de 24% de 9-
v
hydroxyphénanthrène et 17% de 1-hydroxypyrène chez N. virens , et 18% de 9-
hydroxyphénanthrène et 20% de 1-hydroxypyrène chez L. variegatus. D'après ces
importantes proportions en métabolites, nous présumons que les métabolites présents dans
les vers sont principalement dus à la métabolisation des HAP parents plutôt qu'à la
bioaccumulation à partir des sédiments où les métabolites étaient présents en faible
concentration.
La réalisation de ces travaux nous aura permis d'atteindre l' ensemble des objectifs
fixés. Nous avons pu mettre au point deux méthodes d' extraction sélective nous permettant
de déterminer les HAP totaux biodisponibles dans les sédiments fortement contaminés par
des alumineries et plus précisément les HAP avec un log Kow <5,8. De plus, l' étude de la
bioaccumulation nous a permis de montrer que la composition et le niveau de
contamination jouaient un rôle sur la biodisponibilité des HAP. On a pu voir que les vers
bioaccumulaient, en proportion, plus de HAP dans les sédiments faiblement contaminés
que dans les fortement contaminés. Cette étude nous a aussi permis de déterminer que les
différents paramètres toxicocinétiques (ingestion, élimination, facteurs de bioaccumulation)
variaient d'un sédiment à l'autre et d'un HAP à l'autre.
VI
ABSTRACT
Traditionally, polycyclic aromatic hydrocarbon (PAH) concentrations In soil and
sediment have been determined after solid/liquid extractions using organic solvents, such as
dichloromethane or hexane/acetone. However, when hydrophobic organic contaminants
(HOCs), such as P AHs, enter soil or sediment, they undergo several loss or transport
processes which modify their behaviour. In addition, sorption-related pro cesses may cause
these chemicals to become increasingly solvent inextractable. The results of this weathering
process are a corresponding decrease in the extractability of P AHs and a decline of their
availability for benthic organisms and their biodegradability by microorganisms. This
phenomenon has been termed an ' aging effect'. A considerable research effort has been
recently devoted to develop reliable chemical methods for the determination of labile P AHs
in soils and sediments.
The aim of the present study was to establish one or more selective chemical methods
allowing the characterization of bioavailable P AH concentrations in sediments with
variable contamination levels and geochemical properties. Moreover, we also established
accumulation temporal pattern of P AHs in those sediments in two worm species to identify
toxicokinetic parameters describing uptake, elimination and bioaccumulation factors.
Thus, three series of experiments have been carried out. The first one permitted to
compare three different solid/liquid extraction techniques (n-butanol (BuOH),
hydroxypropyl-~-cyc1odextrin (HPCD), and a solution of surfactant Brij700 (B700)) and
vu
the second study to compare the efficiency of two passive samplers, polyoxymethylene
(POM) strips and polydimethylsiloxane (PDMS) silicon tubing. Results of these two
studies were compared with 28-day uptake experiments by Nereis virens for marine
sediments and Lumbriculus variegates for freshwater sediments. In addition, results of the
second study were compared to the previously measured B700 liquid/solid extraction. The
third experiment determined the uptake temporal patterns of the different PARs by worms
(N. virens and L. variegatus) in the different studied sediments as weIl as the organism
ability to metabolize phenathrene and pyrene by the determination of corresponding
hydroxy-PARs (9-hydroxyphenanthrene and I-hydroxypyrene).
Results obtained during the two first studies show that strong differences were
observed in quantities and proportions of PARs obtained using different techniques for
assessing the bioavailability of PARs in sediments. Our first study shows that BuOR
extracted much more PARs than worms can bioaccumulate in their tissues, and where
RPCD underestimated the bioavailability of PARs especiaIly for low molecular weight
PARs. On the other hand, use of the surfactant B700 revealed a good prediction capability
for PARs in highly contaminated sediments. Results of the second study show that PDMS
overestimated the availability of PARs in ail studied sediments (low and highly
contaminated ones), whereas the POM method provided results quite similar to the
solid/liquid extraction using high molecular weight Brij®700. Indeed, bioavailability of
total PARs was correctly predicted by POM and B700 in highly contaminated aluminum
smelter sediments. Rowever, a closer examination of individual PAR results indicated that
both techniques overestimated the availability of large molecules with log Kow >6
Vlll
suggesting the presence of a biological mechanism limiting uptake of larger P AHs which
seem to be related to the molecular size of the compounds or to the steric congestion.
The results of the third study showed that Nereis virens and Lumbriculus variegatus
accumulated sediment-associated PAHs rapidly and achieved apparent steady-state within
seven days (168 h) for aIl studied sediment samples. In addition, the uptake clearance rate
constants of P AHs were lower for highly than for low contaminated sediments (10-2-10-6
and 10-1_10-4 g dry sediment g-I wet organism h- I, respectively), whereas the elimination
rate constants of P AHs remained in same order of magnitude for both type of sediments.
Moreover, this study showed that log of the uptake clearance rate constant (log ks)
decreased with increasing hydrophobicity of PAHs, expressed by the octanol-water
partition coefficient (log Kow) for aIl studied sediment samples. The negative relationship
between log ks and log Kow suggests that the desorption rate of the compounds from the
sediment was an important governing factor in the accumulation by the worms. The
determination of the relative proportion of phase l metabolites toward total phenanthrene
and pyrene in worm tissues after a 28-day exposure indicated the presence of 24% of 9-
hydroxyphenanthrene and 17% of I-hydroxypyrene in N. virens, and 18% of 9-
hydroxyphenanthrene and 20% of 1-hydroxypyrene in L. variegatus. According to these
high proportions of metabolites, we assumed that the metabolites present in worms are
mainly due to metabolization of the parent P AH instead of bioaccumulation from sediment
where metabolites were present in low concentrations.
AlI the fixed objectives have been achieved by the realisation of these experiments.
lndeed, we succeeded to develop two non-exhaustive extraction methods in order to
IX
determine bioavailable P AHs in aluminium smelter highly contaminated sediments and
more precisely PAHs with log Kow <5,8. In addition, the bioaccumulation study showed
that composition and contamination level played a key role on PAH biopavailability. We
also showed that worms bioaccumulated, in proportion, more P AHs in low contaminated
sediments than in highly contaminated ones. This study allowed us to establish that the
toxicokinetic parameters (uptake, elimination, bioaccumulation factors) varied from a
sediment to an other and from a P AH to an other.
x
TABLE DES MATIÈRES
REMERCIEMENTS
RÉSUMÉ Il
ABSTRACT VI
TABLE DES MATIÈRES x
LISTE DES TABLEAUX XIV
LISTE DES FIGURES XVll
INTRODUCTION GÉNÉRALE 1
1. LES HYDROCARBURES AROMATIQUES POLYCYCLIQUES (HAP) 2
1.1. STRUCTURE CHIMIQUE 2
1.2. LES VOIES D' ENTRÉE DES HA? DANS LES ENVIRONNEMENTS ESTUARIENS ET
MARINS
1.3. LES HA? DANS L' EAU
1.4. LES HA? DANS LES SÉDIMENTS
1.5. LES HA? DANS LE BIOTE
2. BIO DISPONIBILITÉ ET TOXICITÉ
4
5
5
6
6
3. SÉQUESTRATION DES CONT AMINANTS ORGANIQUES HYDROPHOBES
PAR DES GÉOSORBANTS
4. BIODISPONIBILITÉ ET BIOACCESSIBILITÉ
12
15
Xl
5. TRA V AUX EFFECTUÉS À L'ISMER 18
5.1. EXTRACTION SÉLECTIVE DES HAP 18
6. OBECTIFS 19
6.1. OBJECTIF GÉNÉRAL 19
6.2. OBJECTIFS SPÉCIFIQUES 19
CHAPITRE 1 21
ENVIRONMENT AL CONTEXT 22
RÉSUMÉ 23
ABSTRACT 25
INTRODUCTION 26
MA TERIALS AND METHODS 29
CHEM ICALS 29
SEDIMENT SAMPLES 31
DCM EXTRACTION 33
BuOH EXTRACTION 33
HPCD EXTRACTION 34
SURF ACTANT EXTRACTION 34
OPTIMIZA TI ON OF B700 EXTRACTION 35
BIOACCUMU LATION STUDIES 38
PAHs fN BIOLOGICAL TISSUES 41
PAHs ANALYSIS 41
DA T A TREA TMENT 42
RESULTS 43
DISCUSSION 57
XIl
EXTRACTABLE PAHs AND SEDIMENT PROPERTIES 57
CHEMICALL y A V AILABILITY v. BIOACCUMULATION 61
CONCLUSION 64
ACKNOWLDGEMENTS 65
REFERENCES 65
CHAPITRE II 73
RÉSUMÉ 74
ABSTRACT 75
INTRODUCTION 76
MA TERIALS & METHODS 78
CHEMICALS 78
SEDIMENTS 78
BIOACCUMULATION STUDIES 81
ANAL YSIS OF WORM TISSUES 81
SEDIMENT CHARACTERIZATION 82
FREELY DISSOL VED CONCENTRATION/SEDIMENT SORPTION EXPERIMENTS 83
PAH EXTRACTION FROM PASSIVE SAMPLERS 83
RESULTS AND DISCUSSION 84
BSAF CALCULATION 84
CORRELATION BETWEEN BSAFs 89
CONCLUSION 95
ACKNOWLDGEMENTS 95
REFERENCES 96
CHAPITRE III 102
RÉSUMÉ
ABSTRACT
INTRODUCTION
MATERIALS & METHODS
CHEMICALS
SEDIMENTS
UPTAKE EXPERJMENTS
ANL YSIS OF WORM TISSUES AND SEDIMENTS
QUANTIFICATION
DA TA TREATMENT
RESULTS
BIOCONCENTRATION OF PAHs IN WORMS
BIOTRANSFORMATION OF PAHs IN WORMS
DISCUSSION
BIOCONCENTRATION OF PAHs IN WORMS
BIOTRANSFORMATION OF PAHs lN WORMS
CONCLUSION
ACKNOWLDGEMENTS
REFERENCES
CONCLUSION GÉNÉRALE & PERSPECTIVES
CONCLUSIONS GÉNÉRALES
CONCLUSION
PERSPECTIVES
BIBLIOGRAPHIE
X 111
103
105
107
109
109
110
III
III
113
114
120
120
124
130
130
131
133
135
135
140
141
146
147
152
LISTE DES TABLEAUX
INTRODUCTION GÉNÉRALE
CHAPITRE 1
Table 1. Sorne chemical and physical properties of the extractants
Table 2. Geochemical properties of the sediment samples
XIV
30
32
Table 3. Reproducibility of the B700 extraction repeated three times on the same marine
sediment
Table 4. Examples ofkinetic data for bioaccumulation of total PAHs (flg g-I)
36
40
Table 5. Results of extractions and bioaccumulation experiments with %XTRAC given in
parentheses. Extractant values followed by a star are significantly different from the
corresponding worm bioaccumulation value. This is an indication of the failure of the
extractant to match the value of total PAHs observed in worms (P>O.01) (mean ± std.
dev. , n = 3) 47
Table 6. Linear correlation parameters obtained when comparing LMW and HMW PAHs
extracted with BUOH, HPCD and B700 with P AHs bioaccumulated in worms for both
low and high contaminated sediments 54
xv
CHAPITRE II
Table 1. Identification of samples for low and highly contaminated sediments. Total PAHs
in sediment (Csed) include: fluorene (FLU), phenanthrene (PHE), fluoranthene (FLT),
pyrene (PYR), benzo[a]anthracene (BaA), chrysene (CHR), benzo[b]fluoranthene
(BbF), benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), dibenzo[a,h]anthracene
(DBA), and benzo[g,h,i]perylene (BPE). TOC = Total organic carbon 80
Table 2. Biota-sediment accumulation factors (BSAF) determined for aIl P AHs in aIl
sediment samples as weIl as ratios between theoretical and empiric BSAFs
(BSAFtheoretica/BSAFworms) and between BSAFs calculated on the basis of
concentrations extracted by the different methods and empirical ones
(BSAF est,POMIBSAF worms, BSAF est,silicOi/BSAF worms and BSAF est,B70o/BSAF worms) 87
CHAPITRE III
Table 1. Concentration of phenanthrene (PHE), 9-hydroxyphenanthrene (9-0H-PHE),
pyrene (PYR), and 1-hydroxypyrene (l-OH-PYR) in sediments and in worms (flg il
w.w.) and percentage ofmetabolite compared to parent PAH. 112
Table 2. Estimated wet weight-based uptake (ks) and elimination (ke) rate constants for the
kinetics of polycyclic aromatic hydrocarbons (P AHs) in polychaetes and oligochaetes.
115
Table 3. Bioaccumulation (BAF) and biota-sediment accumulation (BSAF) factors for the
selected polycyclic aromatic hydrocarbons. 117
XVI
Table 4. Estimated wet weight-based uptake (k,), metabolization (k2) and elimination (k3)
rate constants for the kinetics of hydroxy polycyclic aromatic hydrocarbons (OH-PAHs)
in polychaetes and oligochaetes. 119
XVll
LISTE DES FIGURES
INTRODUCTION GÉNÉRALE
Fig. 1. Structures et noms des 12 HAP étudiés: (1) Fluorène (2) Phénanthrène; (3)
Anthracène; (4) Fluoranthène; (5) Pyrène; (6) Benz[a]anthracène; (7) Chrysène; (8)
Benzo[b]f1uoranthène; (9) Benzo[k]f1uoranthène; (10) Benzo[a]pyrène; (11)
Dibenz[ a,h ] anthracène; (12) Benzo[g,h,i]pérylène. 3
Fig. 2. Modèle conceptuel du domaine du géosorbant. Les lettres encerclées se réfèrent aux
mécanismes de sorption décris précédemment. Le domaine du géosorbant inclus des
différentes formes de matière organique adsorbante (MOA), de carbone particulaire
provenant de résidus de combustion tel que les suies, et de carbone anthropique incluant
les phases liquides non-aqueuses (PLNA) (traduit de Luthy et al. , 1997). 14
Fig. 3. Diagramme conceptuel illustrant les fractions biodisponibles et bioaccessibles d'un
contaminant dans un sol ou sédiment selon sa localisation physique (traduit de Semple
et al. , 2004). (A: Composé sorbé (rapidement réversible) (biodisponible ou
bioaccessible: lié temporairement); B: Composé sorbé (lentement/très lentement
réversible) (bioaccessible: lié temporairement) ; C: Composé bioaccessible (lié
physiquement); D: Composé séquestré (non bioaccessible); E: Composé biodisponible)
(traduit de Semple et al. , 2004). 17
XVlll
CHAPITRE 1
Fig. 1. Extraction efficiency of P AHs from marine sediment with a molarity gradient of
commercial surfactant B700. The optimum concentration has been determined at the
beginning of the plateau. 37
Fig. 2 . Distribution pattern of PAHs in four typical samples used to compare extraction
methods and bioaccumulation using worms. 45
Fig. 3. Comparison of extraction yields and worm bioaccumulation of PAH groups in
highly contaminated sediments. Amounts of P AHs are normalized to the sum of P AHs
extracted with each method. 49
Fig. 4. Comparison of extraction yields and worm bioaccumulation of PAH groups in low
contaminated sediments. Amounts of P AHs are normalized to the sum of P AHs
extracted with each method. 51
Fig. 5. Low (LMW) and high (HMW) molecular weight PAHs extracted by BuOH (a),
HPCD (b) and B700 (c) as a function of bioaccumulated P AHs in worms over 28 days.
The dotted lines represent a hypothetical 1: 1 correlation. 53
Fig. 6. Relationship between PAH octanol-water partitioning coefficient (Kow) and the
ability to predict PAH bioavailability using B700 availability methodology. 56
CHAPITRE II
Fig. 1. Mean of BSAFest,B7oo (A), BSAFest,poM (B), and BSAFest,silicon (C) as a function of
mean BSAFworms over 28 days for low and highly contaminated sediments (left panels).
XIX
BSAF est,B700 (D), BSAF est,POM (E), and BSAF est,silicon (F) of individual P AH as a function
of BSAFworms over 28 days for low contaminated sediments (right panels). The dotted
lines represent a hypothetical 1: 1 correlation. Scrubber sample is identified with letter s.
91
Fig. 2. Relationship between PAH octanol-water partitioning coefficient (log Kow) and the
ability to predict PAH BSAF using B700 and POM methodology. The studied PAHs
are: fluorene (FLU), phenanthrene (PHE), fluoranthene (FL T), pyrene (PYR),
benzo[a]anthracene (BaA), chrysene (CHR), benzo[b]fluoranthene (BbF),
benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), dibenzo[a,h]anthracene (DBA),
and benzo[g,h,i]perylene (BPE). 94
CHAPITRE III
Fig. 1. U ptake kinetics of fluorene ( . ), phenanthrene (0), fluoranthene ( ... ), pyrene ( . ),
bz[a]anthracene + chrysene (v), benzo[bk]fluoranthene (. ), benzo[a]pyrene (0), and
benzo[g.h.i]perylene (. ) in worms during the exposure period in B Lagoon (A),
Scrubber (B), SLR Downstream (C), EGSL04-07 (D), SLR Upstream (E), and Hospital
Beach (F) sediments. The curve corresponds to the nonlinear fit of the data using the
model expressed in Eq. 1. Note different scales for PAH concentrations. 121
Fig. 2. (Top) Plot of the log of uptake clearance rate constants, log ks, versus the log of the
hydrophobicity expressed by the octanol/water partition coefficient, log Kow, for highly
(A) and low (B) contaminated sediments. (Bottom) Plot of the log of elimination rate
xx
constants, log ke, versus the log of the hydrophobicity expressed by the octanol/water
partition coefficient, log Kow, for highly (C) and low (D) contaminated sediments. 123
Fig. 3. The ratio of 9-hydroxyphenanthrene to phenanthrene (. ) and bioaccumulation
kinetic of phenanthrene (0) during the exposure period in B Lagoon (A), Scrubber (B),
SLR Downstream (C), EGSL04-07 (D), SLR Upstream (E), and Hospital Beach (F)
sediments. Note different scales for phenanthrene concentration. 125
Fig. 4. The ratio of I-hydroxypyrene to pyrene (. ) and bioaccumulation kinetic of pyrene
(0) during the exposure period in B Lagoon (A), Scrubber (B), SLR Downstream (C),
EGSL04-07 CD), SLR Upstream CE), and Hospital Beach (F) sediments. Note different
scales for pyrene concentration. 127
Fig. 5. Plot of the elimination rate constants Ck3) versus the metabolization rate constants
(k2) for 9-hydroxyphenanthrene in ail (A), highly (B), and low contaminated sediments
CC) and for I-hydroxypyrene in ail (D), highly CE), and low contaminated sediments
CF) . 129
INTRODUCTION GÉNÉRALE
2
1. Les hydrocarbures aromatiques polycycliques (HAP)
Les hydrocarbures aromatiques polycycliques (HAP) sont les xénobiotiques
organiques les plus présents dans les environnements estuariens et marins. Ceux-ci sont
devenus extrêmement importants à cause de leurs effets cancérigènes, mutagènes et
tératogènes potentiels sur les organismes aquatiques et sur l'homme (Fetzer, 2000).
1.1. Structure chimique
Les hydrocarbures aromatiques polycycliques sont un groupe de composés constitués
d'atomes d'hydrogène et de carbone formant de deux à sept cycles benzéniques dans des
arrangements linéaires, angulaires ou en grappes avec des groupes insaturés possiblement
attachés à un ou plusieurs cycles (Fetzer, 2000). Ces composés vont du naphtalène (CIOHS,
deux cycles) au coronène (C24H I2 , sept cycles) (Fig. 1). En général, les HAP ont une faible
solubilité dans l'eau, un haut point de fusion et d'évaporation et une faible pression de
vapeur. Quand la solubilité diminue, les points de fusion et d'évaporation augmentent, et la
pression de vapeur diminue avec l'augmentation du volume moléculaire qui est directement
fonction du nombre de cycles (Fetzer, 2000).
3
Fig. 1. Structures et noms des 12 HAP étudiés: (1) Fluorène (2) Phénanthrène; (3)
Anthracène; (4) Fluoranthène; (5) Pyrène; (6) Benz[a] anthracène; (7) Chrysène; (8)
Benzo[b]f1uoranthène; (9) Benzo[k]f1uoranthène; (10) Benzo[a]pyrène; (11)
Di benz [ a,h ] anthracène; (12) Benzo [g,h,i ]péry lène.
(1)
(4)
(7)
(10)
~
(2)
# #
(5)
(8)
(11)
(3)
~
~
(6)
(9)
(12)
4
1.2. Les voies d'entrée des HAP dans les environnements estuariens et marins
Les routes principales d'entrée des HAP dans les environnements estuariens et marins
incluent les apports atmosphériques et hydriques provenant principalement d'activités
anthropiques (les effluents industriels et urbains, les déchets d'incinération, les accidents
pétroliers (le pétrole brut contient de 0,2 à 7 % de HAP)) (Lima et al. , 2005), la production
d'asphalte et la combustion des combustibles fossiles (Dabestani et Ivanov, 1999; Oros et
Ross, 2004; Yunker et al. , 2002), la diagenèse de la matière organique dans les sédiments
anoxiques (Lima et al. , 2005) ou encore de produits et de processus naturels tels que les
feux de forêts et les émissions volcaniques (Bowerman, 2004; Lu, 2003).
Presque tous les HAP provenant de l'atmosphère sont associés à la matière
particulaire aérienne et aux aérosols à cause de leurs coefficients de partage octanolleau
(Kow) relativement élevés, ce qui dénote un important potentiel d' adsorption sur les
matières particulaires en suspension dans l'air et dans l'eau (Law et al. , 2002). La pluie et
la neige représentent le principal processus atmosphérique responsable du flux de HAP
dans les océans mondiaux (Dickhut et al., 2000; Tsai et al. , 2002) pour trouver leur
destination finale dans les sédiments (Christensen et Bzdusek, 2005). Ainsi , le sédiment
devient un puits pour les HAP à cause de leur relative immobilité quand ils sont non-
perturbés, et de leur persistance (demi-vie de 0.3 à 58 ans pour le benzo[a]pyrène) (Ghosh
et al. , 2003; Gulnick, 2000).
5
1.3. Les HAP dans l'eau
Les eaux de surface comme les rivières et les eaux côtières peuvent être fortement
polluées par les HAP en raison des industries et des ports. Les concentrations totales en
HAP dans les eaux de rivières traversant des régions fortement industrialisées peuvent
atteindre plus de 1 ° Ilg/L de HAP totaux. Les eaux de rivières non-polluées et l'eau de mer
contiennent moins de 0,1 Ilg/L de HAP totaux (World Health Organization, 1998).
Les concentrations des HAP sont plus faibles dans la colonne d'eau que dans le biote
et les sédiments, due en partie à la faible solubilité dans l'eau des HAP. La matière
organique dissoute ou colloïdale, telle les acides humiques et fulviques, dans l'eau de mer
peut agir comme un solvant pour les HAP. Plus le nombre de cycles aromatiques ou la
masse moléculaire des HAP augmente, plus la solubilité diminue. Par exemple, la solubilité
dans l'eau du naphtalène, un HAP à deux cycles, est d'environ 30 mg/L, alors que celle des
HAP à cinq cycles varie entre 0,5 et 5,0 Ilg/L (Mackay et al. , 1991).
1.4. Les HAP dans les sédiments
Les HAP sont facilement adsorbés à la surface des particules à cause de leur
hydrophobicité, leur faible solubilité dans l'eau, leur relativement faible pression de vapeur
et leur aromaticité (Pignatello et Xing, 1996). La rapidité à laquelle les HAP s'adsorbent sur
la matière particulaire en suspension entraîne une concentration plus importante de ces
contaminants dans les sédiments des fonds marins que dans la colonne d'eau (généralement
par un facteur 1000 ou plus) (Dungavell, 2005). Les sédiments estuariens et marins
6
deviennent donc un réservoir majeur de HAP et une source continue de contamination pour
les communautés biotiques (Ghosh et al. , 2003; Gulnick, 2000).
Étant donné la proximité des sources anthropiques et de la nature hydrophobe des
HAP, les dépôts marins côtiers ont des concentrations significativement élevées par rapport
aux autres régions marines, tels que les sédiments des marges ou des abysses. Les zones
recevant le drainage des centres industrialisés peuvent avoir des niveaux de HAP de 10 000
Ilg/g ou plus dans les sédiments. Dans les régions éloignées des activités anthropiques, les
valeurs des HAP totaux dans les sédiments sont souvent de l'ordre du Ilg/g (Gulnick, 2000).
1.5. Les HAP dans le biote
Les concentrations des HAP dans les orgamsmes aquatiques sont extrêmement
variables selon les sites et les espèces aquatiques. Les valeurs rapportées sont comprises
entre 0,01 et 5 000 Ilg/kg de poids sec (McElroy et al. , 1989). Les concentrations élevées
dans les organismes marins se présentent souvent dans les zones recevant des décharges
chroniques d'hydrocarbures (McElroy et al., 1989).
Les organismes déposivores peuvent ingérer les contaminants organiques à partir de
l' eau les entourant, de l'eau porale par respiration, filtration ou par adsorption directe, et
aussi à partir des particules sédimentaires ingérées par désorption et adsorption en présence
de fluides digestifs (Weston et al., 2000).
2. Biodisponibilité et toxicité
La biodisponibilité d'un composé chimique est une mesure de son accessibilité au
biote dans l'environnement. Elle est définie ici comme la fraction du contaminant qui est en
7
fin de compte disponible pour l'ingestion, l 'accumulation ou l'assimilation par un
organisme. C'est un facteur important contrôlant l' ingestion de contaminants liés au
sédiment dans les organismes aquatiques, et le transfert des contaminants dans la chaîne
alimentaire. C'est aussi un facteur critique dans le succès des techniques biologiques de
remédiation. La biodisponibilité des contaminants hydrophobes tels les HAP est déterminée
par les interactions complexes de différents facteurs abiotiques et biotiques, tels que les
caractéristiques du composé, les caractéristiques du sédiment, et les caractéristiques
biologiques des organismes (Lu, 2003).
Les facteurs fondamentaux affectant la disponibilité des contaminants sont la matière
organique des sols et sédiments (Kukkonen et al. , 2003; Oleszcuk et Baran, 2004; Rust et
al. , 2004a; Shor et al., 2003 ; Thorsen et al. , 2004), le pH, la nanoporosité, le pourcentage en
argile (Nam et Alexander, 1998), l' hydrophobicité (Nam et Alexander, 1998), et la capacité
d 'échange en cations (Chung et Alexander, 2002). La matière organique inclue le matériel
de décomposition des plantes et animaux aussi bien que l'humine provenant de racines.
Étant donné l'affinité importante des composés organiques à la matière organique des sols
et sédiments, celle-ci réduit la disponibilité de ces composés dans les sols et sédiments. Des
études évaluant la biodisponibilité et la séquestration des HAP dans les sédiments ont
rapporté que les concentrations en contaminants (Chung et Alexander, 1999a), le taux
d'humidité (Kottler et Alexander, 2001), la remise en suspension (Talley et al., 2002; White
et al., 1999a), la présence d'autres HAP (White et al., 1999a; 1999b) et la durée de contact
entre le polluant et le sédiment (Macleod et Semple, 2000) sont d' autres facteurs
significatifs influençant la biodisponibilité du contaminant.
8
Des études ont indiqué que les sols et sédiments avec un taux de matière organique
élevé influence la disponibilité. Le contenu en HAP extraits avec du n-butanol a été relié au
type de boue de traitement des eaux usées (Oleszczuk et Baran, 2004). Talley et al. (2002)
ont rapporté que la réduction en HAP dans la fraction argile/silt se corrélait bien avec la
dégradation des HAP et l'accumulation par des vers de terre. Nam et al. (1998) ont rapporté
que le phénanthrène extrait du sol avec du n-butanol était proportionnel au taux de carbone
organique dans le sol. La dissipation initiale des contaminants peut être attribuée au taux
d 'humidité du sol, à la température et à d'autres facteurs environnementaux.
Des études récentes supportent le phénomène de vieillissement, suggérant que les
composés organiques persistant dans les sols et sédiments deviennent moins disponibles à
l'ingestion par les organismes, deviennent moins toxiques, et sont moins disponibles pour
la biodégradation par les microorganismes. Ces observations indiquent que les molécules
des contaminants sont prisonnières dans des micro-sites qui ne sont pas faciles d'accès au
microorganismes. Chung et Alexander (1999b) ont montré qu'il y avait une relation entre le
carbone organique et des pores de diamètre 0,1-10 J-lm et entre le contenu en argile et des
pores <102 nm. Ainsi, la fraction biodisponible des contaminants peut ne pas être
déterminée précisément par des analyses chimiques déterminant les concentrations totales
en polluant (Alexander, 2000).
Les premières informations sur l'effet du vieillissement et sur la biodisponibilité
proviennent d'études sur les concentrations de pesticides (dichlorodiphényltrichloroéthane
(DDT), aldrine, dieldrine, heptachlore et chlordane) dans les sols en culture mesurées
pendant de longues périodes et de mesures de toxicité de ces pesticides pour les invertébrés
9
et les plantes (Decker et al. , 1965; Gilbert et Lewis, 1982; Korschgen, 1971 ; Lichtenstein et
al. , 1960; Lichtenchtein et Schulz, 1965; Onsager et al. , 1970; Wingo, 1966). Bien que la
disparition initiale d'un pesticide soit partiellement le résultat de sa volatilisation ou de sa
dégradation abiotique aussi bien que de sa biodégradation par les microorganismes, le fait
que cette disparition faiblisse au fur et à mesure des années indique que ces insecticides
sont devenus faiblement disponibles aux microorganismes indigènes. Les premières études
toxicologiques ont aussi démontré une relation temporelle avec la diminution de la
biodisponibilité. Par exemple, des dosages biologiques de la toxicité aiguë sur la mouche
Drosophila melanogaster simultanément avec des déterminations chimiques ont donné des
résultats similaires peu de temps après l'application de lindane à un sol, mais une partie
importante de cet insecticide restant dans le sol après 22 mois n'affectait plus de façon
détectable les mouches (Edwards et al. , 1957).
Des études récentes ont montré que des composés organiques présents depuis de
nombreuses années dans des sols sont moins biodisponibles que ceux fraîchement ajoutés à
ces mêmes sols. Morrison et al. (2000) ont trouvé que du DDT, du
dichlorodiphényldichloroéthylène (DDE) et du dichlorodiphényldichlorométhane (DDD)
d ' un site de déchets (contaminé 30 ans auparavant) étaient approximativement
biodisponibles à 30% pour des vers. La diminution de disponibilité pour de la simazine, des
HAP et du 1,2-dibromethane vieillis a aussi été démontrée (Erickson et al. , 1993 ;
Weissenfels et al. , 1992).
Le vieillissement est toxicologiquement significatif parce que l' assimilation, la
toxicité aiguë et chronique des composés dangereux diminuent pendant qu ' ils persistent et
10
deviennent de plus en plus séquestrés avec le temps (Alexander, 2000). La toxicité peut être
moindre que celle anticipée même quand elle est basée sur des extractions vigoureuses des
sols et sédiments (Erickson et al. , 1993; Meier et al., 1997; Peijnenburg et al., 2000). Bien
que le vieillissement réduit l'exposition et ainsi la toxicité et le risque, il ne les élimine pas.
Des composés tels que le TCDD, BPC et PBB qui ont persisté pendant de longues périodes
sont toujours biodisponibles pour les mammifères. Des sols traités au DDT et au chlordane
étaient toujours toxiques pour des vers même après une période de vieillissement
(Alexander, 2000). Du phénanthrène et du pyrène vieillis dans un sol montrèrent une
diminution de la biodisponibilité aux microbes ainsi qu 'une toxicité pour des vers avec le
temps (Chung et Alexander, 1999a). Il est important de noter que même si les composés
vieillissent, une fraction du composé reste biodisponible et il a été montré que cette fraction
s'accumule dans des insectes, plantes et bactéries.
Le gestionnaire de l'environnement est face à un dilemme majeur car l'importance de
la réduction de la biodisponibilité résultant du vieillissement est différente pour le même
composé dans différents sols, pour différents composés dans le même sol et pour des durées
différentes d'exposition du composé dans le sol. Comment peut-on évaluer le degré
d'exposition et prédire le risque d'un composé vieilli? Les biotests sont un moyen évident
de faire des estimations, mais la précision de ces méthodes biologiques est parfois
inadéquate pour les besoins de la réglementation. Certains prennent beaucoup de temps et
sont chers. Une alternative serait un test chimique ou physique dont les résultats seraient
clairement corrélés avec les résultats des biotests.
Il
Un effort considérable a été récemment déployé pour le développement de méthodes
d'extraction afin de déterminer les concentrations biodisponibles ou labiles des
contaminants dans les sols et sédiments. Afin d'évaluer la fraction biodisponible des
contaminants dans les sols et sédiments de nombreuses méthodes d'extraction ont été
élaborées. Des méthodes d'extraction utilisant des échantillonneurs passifs ("passive
sampler") tels que le Tenax, XAD, des membranes semiperméables, des disques de silice
octadecyle modifié, du polyoxyméthylène (POM), des tubes de silicone (Akkanen et al.,
2001; Cornelissen et al., 1998; Cornelissen et al. , in press; Cuypers et al., 2002; Krauss et
Wilcke, 2001; Richardson et al. , 2003; Rust et al., 2004b; Tang et al. , 2002; van der Wal et
al., 2004; Zimmerman et al. , 2004), des gaz hautement pressurisés (Hawthorne et
Grabanski, 2000; Librando et al., 2004; Loibner et al., 2000; Szolar et al., 2001; Szolar et
al., 2004), et l'oxydation au persulfate (Cuypers et al., 2000) ont été proposées.
Récemment, des solutions aqueuses de cyc10dextrines (Cuypers et al. , Reid et al. , 2000;
2002; Swindell et Reid, 2006) et de tensioactifs (Chang et al., 2000; Cuypers et al., 2002;
Fava et al., 2004; Tang et al., 2002) ont été testées sur des sols et sédiments contaminés
ainsi que des systèmes multi-colonnes (Zhao et Voice, 2000) et des extractants chimiques
sélectifs (Tang et al., 2002) tel que le n-butanol (Alexander et Alexander, 2000; Liste et
Alexander, 2002; Oleszczuk et Baran, 2004).
Les résultats des analyses par de telles procédures ont été corrélés avec la
biodisponibilité pour différents organismes vivants. Lors de l'analyse de telles méthodes
dites douces, les auteurs se sont aperçus que la diminution de la biodisponibilité en fonction
du temps était accompagnée d'une chute de la quantité extraite de polluants.
12
Ces observations suggèrent que de telles méthodes d'extraction sélectives pourraient
servir de base pour un test de substitution afin de déterminer la biodisponibilité.
3. Séquestration des contaminants organiques hydrophobes par des géosorbants
Au cours des années 1990, le professeur Luthy et collaborateurs ont proposé un
modèle conceptuel pour décrire la séquestration des composés hydrophobes par les
géosorbants. Les mécanismes à la base de la séquestration des contaminants organiques
hydrophobes sont complexes et font appel à une succession d'étapes à partir d'une
adsorption à la surface des particules suivie d'une absorption en profondeur que l'on peut
schématiser ainsi (adapté de Luthy et al., 1997).
A. Absorption rapide dans la matière orgamque amorphe ou dans un liquide
organique inter-particulaire: mécanisme plus favorable aux molécules polaires,
molécules peu ou pas accessibles aux organismes, faiblement extraites.
B. Absorption lente dans la matière organique vieillie et dense: molécules peu ou
pas accessibles aux organismes, séquestration complète.
C. Adsorption rapide sur les surfaces organiques mouillées par l'eau interstitielle:
interactions des régions polaires entre molécules adsorbées et l'absorbant,
biodisponibles et faciles à extraire.
D. Adsorption rapide sur les surfaces minérales mouillées par l'eau interstitielle: les
molécules adsorbées sont disponibles à la biodégradation et faciles à extraire.
13
E. Adsorption dans les mlcropores de la matrice minérale (argile): molécules
probablement peu accessibles à la biodégradation mais demeurent extractibles
avec des solvants organiques.
La Fig. 2 schématise les types de mécanisme appliqués aux divers géosorbants d'un sol.
14
Fig. 2. Modèle conceptuel du domaine du géosorbant. Les lettres encerclées se réfèrent aux
mécanismes de sorption décrits précédemment. Le domaine du géosorbant inclus des
différentes formes de matière organique adsorbante (MOA), de carbone particulaire
provenant de résidus de combustion tel que les suies, et de carbone anthropique incluant les
phases liquides non-aqueuses (PLNA) (traduit de Luthy et al. , 1997).
lVIOA mnol11ltr rucallsult'
RESIDU DE COMBUSTION
Risidu d~ combusliou, ~x.: suirs
1 1
1
Micropores ----'lW/.
Mésopores --1--+11 ....
EAU OU GAZ DANS DES 1 MACROPORES 1
.......... "'-"'''''' ... .._--MOA
GEOSORBANT
MOA tltus~
l"lOA lImol]lh~
PLNA ,1rillirs \
15
4. Biodiponibilité et bioaccessibilité
De nombreuses définitions de la biodisponibilité ont été données depuis les 2
dernières décennies (Alexander, 2000; Herrchen et al., 1997; Klaasen, 1986; Kramer et
Ryan, 2000; Ruby et al., 1996; Spacie et al., 1995; Van Leeuwen et Hermens, 1995). Avoir
autant de définitions différentes crée une confusion chez les scientifiques
environnementaux. Ainsi, Semple et al. (2004) ont proposé une définition pour la
biodisponibilité et la bioaccessibilité:
la biodisponibilité est ce qui est librement disponible pour traverser la membrane
cellulaire d'un organisme à partir du média où vit un organisme à un moment
donné (Fig. 3). Une fois que le transfert à travers la membrane a eu lieu, un
stockage, transformation, assimilation, ou dégradation peut se produire dans
l' organisme.
la bioaccessibilité est ce qui est disponible pour traverser la membrane cellulaire
d'un organisme à partir de l'environnement, si l'organisme a accès au
contaminant. Cependant, le contaminant peut être soit physiquement déplacé par
l'organisme, soit devenir biodisponible après une certaine période. Dans ce
contexte, physiquement déplacé peut référer à un contaminant qui est séquestré
dans la matière organique et ainsi est indisponible à un moment donné ou qui
occupe un espace de l'environnement différent de l'organisme (Fig. 3). Les
contaminants peuvent devenir disponible suivant une libération rapide à partir
d'un amalgame labile, ou l' organisme peut bouger et entrer en contact avec le
16
contaminant. Alternativement, une libération peut intervenir bien après (années
ou décennies) et rendre le contaminant bioaccessible. En résumé, la
bioaccessibilité englobe ce qm est réellement biodisponible et ce qui est
potentiellement biodisponible.
Il est bien connu que la portion d'un contaminant qui est, soit biodisponible, soit
bioaccessible, dans un sol ou sédiment peut différer de façon importante selon les
organismes. Un avantage clair de ces définitions est leur multi-fonctionnalité, puisque ces
définitions peuvent s'appliquer aux contaminants étant disponibles ou accessibles aux
microorganismes, champignons, plantes, invertébrés, et animaux supérieurs par passage à
travers la membrane de l'organisme en question. Ceci pourra être la membrane cellulaire
d'une bactérie alors que chez les vers, par exemple, ceci englobe l'ingestion par la peau et
le tractus gastrointestinal.
La distinction entre biodisponibilité et bioaccessibilité force les praticiens à
considérer ce qu' ils mesurent réellement avec les méthodes biologiques et chimiques, qui
sont développées afin de déterminer la fraction biodisponible.
17
Fig. 3. Diagramme conceptuel illustrant les fractions biodisponibles et bioaccessibles d'un
contaminant dans un sol ou sédiment selon sa localisation physique (traduit de Semple et
al. , 2004). (A: Composé sorbé (rapidement réversible) (biodisponible ou bioaccessible: lié
temporairement); B: Composé sorbé (lentement/très lentement réversible) (bioaccessible:
lié temporairement); C: Composé bioaccessible (lié physiquement); D : Composé séquestré
(non bioaccessible); E: Composé biodisponible) (traduit de Semple et al. , 2004).
A Racine
Vers
18
5. Travaux effectués à l'ISMER
5.1. Extraction sélective des HAP
Des techniques d' extraction sélective ont été au centre des récentes recherches dans
l'espoir qu'elles puissent atteindre la fraction labile ou biodisponible. Comme on a pu le
voir précédemment, plusieurs auteurs ont déjà proposé des protocoles d'extraction sélective
des hydrocarbures aromatiques polycycliques (Cuypers et al. , 2000; Librando et al. , 2004;
Oleszczuk et Baran, 2004; Reid et al., 2000; Swindell et Reid, 2006; Szolar et al. , 2004).
Malheureusement, la plupart de ces techniques ne miment que faiblement les processus
inhérents à la biodisponibilité. De plus, celles-ci négligent le fait que les bactéries sont
capables de sécréter des exoenzymes et des biopolymères, ayant des propriétés
tensioactives, pour aider la dissolution partielle du matériel organique particulaire et aussi
permettre le passage à travers leur paroi cellulaire. Une méthode récemment développée
dans notre laboratoire (Barthe, 2002) utilise un tensioactif à haut poids moléculaire (Brij®
700) afin d'extraire la fraction biodisponible des molécules hydrophobes adsorbées sur la
matrice sédimentaire. L'utilisation de tensioactifs à des concentrations proches ou
supérieures à la concentration micellaire critique (CMC) permet de solubiliser des
composés non-ioniques ajoutés dans un mélange d'eau et de sol (Liu et al. , 1991). Notre
méthode repose sur l' hypothèse que l'utilisation du tensioactif mimerait l' action des
bactéries capables de sécréter des exoenzymes et des biopolymères possédant des
propriétés tensioactives qui permettent une dissolution du matériel organique ainsi que la
diffusion des molécules à travers leur membrane cellulaire (Barthe, 2002).
19
6. Objectifs
Comme on a pu le voir précédemment, les études réalisées sur les mécanismes et la
cinétique de séquestration des HAP dans les sédiments ne l'ont été que sur quelques
échantillons à la fois. Notre étude permettra d'avoir une idée plus précise sur les liens qu'il
peut y avoir entre ceux-ci et les caractéristiques physico-chimiques des sédiments puisque
nous allons étudier de nombreux sédiments lacustres et marins de provenance, de
composition et de concentration totale en HAP différentes .
6.1. Objectif général
La présente étude a pour but de caractériser les mécanismes et la cinétique de
séquestration des HAP dans les sédiments lacustres et marins ayant des caractéristiques
géochimiques et environnementales très différentes ainsi que de déterminer la
bioaccumulation des HAP afin de la corréler avec les données de séquestration obtenues.
6.2. Objectifs spécifiques
Afin d'atteindre efficacement l'objectif général présenté, la présente étude a été
divisée en trois objectifs spécifiques. Les travaux d'échantillonnage, d' observation,
d ' analyse et d' interprétation seront ainsi répartis en fonction de chacun de ces objectifs:
1. Développer et intégrer les notions de séquestration des hydrocarbures aromatiques
polycycliques (HAP) dans les sédiments marins et lacustres. Pour ce faire , nous
allons étudier différentes méthodes d'extractions sélectives sur des sédiments de
provenance, de composition géochimique et de concentration totale en HAP
20
totalement différentes afin de déterminer si la composition et la contamination de
ces sédiments peuvent jouer un rôle sur la séquestration des HAP.
2. Déterminer la bioaccumulation des HAP dans les sédiments marins et lacustres en
fonction de leur nature géochimique. Pour ce faire , des études de bioaccumulation
seront effectuées à l'aide de polychètes et d'oligochètes afin de déterminer si la
composition et le niveau de contamination des sédiments jouent un rôle sur la
biodisponibilité des HAP.
3. Déterminer les patrons temporels d' accumulation des HAP pour plusieurs
échantillons de niveau de contamination variable dans deux espèces de vers en
déterminant les paramètres toxicocinétiques décrivant l'ingestion, l'élimination et
les facteurs de bioaccumulation. De plus, l'habilité des organismes à métaboliser le
phénanthrène et le pyrène sera examinée par la détermination d'hydroxy-HAP
sélectionnés.
CHAPITRE 1
COMP ARING BULK EXTRACTION METHODS FOR CHEMICALLY
A V AILABLE POLYCYCLIC AROMATIC HYDROCARBONS WITH
BIOACCUMULATION IN WORMS
M. Barthe and É. Pelletier
(Environmental Chemistry, 2007, 4: 271-283)
22
Environmental context
Determining the bioavailability of organic contaminants in sediments is a critical step
ln assessing the ecological risks of contamination in aquatic ecosystems. Standardised
sediment bioaccumulation tests using benthic organisms are often performed to determine
the relative bioavailability of sediment contamination. Unfortunately biological methods
are time consuming, expensive and organisms are often difficult to maintain in good health
in a laboratory exposure system. Contradictory results have been reported in the last decade
and factors that affect the behaviour of extractants need to be examined for a large range of
sediments. A study was conducted to determine the bioavailability of polycyclic aromatic
hydrocarbons (P AHs) in sediment using worms and to compare the uptake by biological
samplers with mi Id solid/liquid extractions when exposed to unspiked low and highly
contaminated marine and freshwater sediments.
23
Résumé
L'objet de cette étude est d'évaluer différentes techniques afin de déterminer la
disponibilité des hydrocarbures aromatiques polycycliques (HAP) dans les sédiments
contaminés. Ce but fut atteint en comparant les résultats d'études sur 28 jours de
bioaccumulation par Nereis virens et Lumbriculus variegatus avec les HAP extraits par
trois méthodes d'extraction non-exhaustives utilisant le n-Butanol (BuOH, 100%), une
solution aqueuse d'hydroxypropyl-~-cyclodextrine (HP CD) et une solution de tensioactif
de Brij700 (B700). Nos résultats montrent l'importance de considérer le niveau en HAP
dans les sédiments ainsi que la taille moléculaire des HAP quand vient le moment de
prédire leur bioaccumulation dans un échantillonneur biologique comme les vers en
utilisant une méthode d'extraction solide/liquide. La solution de tensioactif B700 a eu du
succès à prédire la bioaccumulation quand elle a été exposée à des sédiments fortement
contaminés (25-5700 j.lg g - 1). Quand des sédiments faiblement contaminés (0,06-1 ,1 j.lg
g - 1) ont été utilisés, le HPCD et le BuOH ont été des meilleurs agents d'extraction pour
estimer la bioaccumulation alors que le B700 apparaît être trop faible pour la majorité des
échantillons. Nos résultats illustrent l'intérêt et les difficultés à trouver un prédicteur
chimique adéquat pour la biodisponibilité des HAP, particulièrement parce que les
concentrations en HAP et les processus de séquestration jouent un rôle déterminant dans la
qualité des résultats. Comme le B700 est bon marché et que les solutions d'extraction sont
faciles à préparer, une procédure d'extraction utilisant ce tensioactif est proposée comme
un prédicteur fiable pour les sédiments fortement contaminés.
24
Mots clés additionnels: biodisponibilité, Brij700, butanol, hydroxypropyl-~-cyclodextrine,
hydrocarbures aromatiques polycycliques (HAP)
25
Abstract
The purpose of this study is to evaluate different techniques for assessmg the
availability of polycyclic aromatic hydrocarbons (P AHs) in contaminated sediments. This
goal was achieved by comparing results from 28-day uptake experiments by Nereis virens
and Lumbriculus variegatus with P AHs extracted by three non-exhaustive extraction
methods using: n-Butanol (BuOH, 100%), an aqueous solution of hydroxypropyl-~
cyclodextrin (HP CD) and a surfactant solution of Brij700 (B700). Our results highlight the
importance of considering both the P AH level in sediments and the molecular size of P AHs
when attempting to predict their bioaccumulation in a biological sampler like worms using
a solidlliquid extraction method. The surfactant B700 solution was quite successful to
predict P AH bioaccumulation when exposed to unspiked highly contaminated sediments
(25-5700 f..lg g- I). When low contaminated sediments (0.06-1.1 f..lg g- I) were used, HPCD
and BuOH were better extractants for estimating bioaccumulation whereas B700 appeared
to be too mild as ex tractant for most samples. Our results illustrate the interest and
difficulties in finding an adequate chemical predictor for PAH bioavailabilty, particularly
because P AH concentrations and sequestration processes play a determining role in the
quality of results . Because B700 is not expansive and extraction solutions are easy to
prepare, an extraction procedure involving this surfactant is proposed as a reliable predictor
for aged highly contaminated sediments.
Additional keywords: bioavailability, Brij700, butanol, hydroxypropyl-~-cyclodextrin,
polycyclic aromatic hydrocarbons (P AHs).
26
INTRODUCTION
Polycyclic aromatic hydrocarbons (P AHs) are pollutants of great environrnental
concern because of their mutagenic and carcinogenic properties. [1 ,2] Traditionally, PAH
concentrations in soil and sediment have been determined after solid/liquid extractions
using organic solvents, such as dichloromethane or hexane/acetone. However, when
hydrophobic organic contaminants (HOCs), such as PAHs, enter soil or sediment, they
undergo several loss or transport processes including volatilisation and leaching, as well as
chemical and biological degradation which modify their behaviour. In addition, sorption-
related processes may cause these chemicals to become increasingly solvent nonextractable
or irreversibly bound within the soil or sediment matrix. The result of this weathering
pro cess with time is a corresponding decrease in the extractability of P AHs. This
phenomenon has been termed an ' aging effect' and is related to a decline of their
availability for uptake by benthic orgamsms and their biodegradability by
microorganisms. [3-5] A considerable research effort has been recently devoted to develop
reliable chemical methods for the determination of labile P AHs. Laboratory methods based
on solid-phase extraction (SPE), [6, 7] persulfate oxidation, [8] solvent extraction[9, ID] and
supercritical fluid extraction[ll] have been proposed. Recently, aqueous solutions of
cyclodextrin[1 2. 13] and surfactants['2, 14] have been tested on contaminated soils and
sediments.
27
In an attempt to compare different methods, Kelsey et al. (9) measured the
extractability of phenanthrene using different mild che mi cal extractants to assess its
bioavailability to earthworms and bacteria. Results suggested that n-butanol (BuOH), a
small linear and polar molecule, was the most appropriate solvent for predicting
bioavailability of phenanthrene to both organisms, but BuOH was unsuccessful to predict
the bioavailability of atrazine. Several other studies using freshly spiked soils came in
support to the BuOH-extraction procedure and showed high correlations between extraction
efficiency and bioavailability of P AHs for biodegradation or bioaccumulation by
worms.[IO, 15, 16]
Cyclodextrins, a family of cyclic oligosaccharides highly soluble in water, were also
considered as extractants because they have a hydrophobic organic cavity in the interior of
the toruS[1 7] that can contribute to the formation of a 1: 1 inclusion complex between the
cyclodextrin macromolecule and an organic moiety. [I 8) Aqueous solutions of cyclodextrin
have been used to dissolve a large range of contaminants of low aqueous solubility. It has
been shown that molecules too large to form l: 1 inclusion complexes with P-cyclodextrin
can forml:2 inclusion complexes, which consist of two macrocYcles and one guestJl 9] The
application of cyclodextrin extraction for the prediction of labile P AHs was first studied by
Reid et al. [1 2,20] So far, only few data with PAHs in field-aged samples have been reported .
Using three dissimilar spiked soils, Swindell and Reid[21] observed that cyclodextrin and
BuOH showed different efficiencies especially for soils with high clay and organic
contents.
28
The application of surfactant extraction for the prediction of chemical lability was
first proposed by Volkering et al. [22] who showed that mineraI oil bioavailability could be
successfully predicted by an extraction with a non-ionic surfactant (Triton X-IOO). Triton
was also used to extract P AHs from two contaminated sediments with high organic matter
content (9.7 to 13.3%) and results compared with hydroxypropyl-~-cyclodextrin (HP CD)
extracts.[l4] Authors observed that Triton X-lOO extracted more high-molecular-weight
P AHs than cyclodextrin and tended to overestimate the bioavailability of P AHs. Among
other surfactants that can be used in a solid/liquid extraction process, our attention was
attracted by poly( oxyethylene)(1 OO)stearyl ether (Brij700), a water soluble non-ionic high-
molecular-weight surfactant, which forms micelles above its critical micelle concentration
(CMC). PAHs and other hydrophobic molecules can be incorporated into these micelles,
which lead to an increase of their aqueous solubility. In addition, a long poly(oxyethylene)
chain (n=lOO) might show sorne similarities with biosurfactants produced by bacteria and
currently found in oil contaminated soils. [23,24]
Most comparisons between extraction techniques mentioned above have been
conducted with spiked soils while paying attention to aging effects by storing treated
samples for weeks and months before running extraction and bioavailability
experiments.[1O,21] Little attention has been paid to freshwater and marine sediments with
low levels of P AHs or to sediments heavily loaded with high temperature combustion
residues.
In an attempt to compare the chemical lability of low and high-molecular-weight
P AHs more or less sequestrated in field sediments, samples from low contaminated St.
29
Lawrence Estuary and from two highly contaminated industrial sites, were extracted with
three non-exhaustive extraction methods using: BuOH aloneYO] an aqueous solution of
HPCD[12,14] and a surfactant solution using high-molecular-weight Brij700 (B700).
Sediments were not spiked with labelled P AHs to avoid possible bias introduced by the
aging process recreated in the laboratory. Total extractable PAHs were determined by
dichloromethane (DCM) extraction. For comparison purposes, the same sediment samples
were used for bioaccumulation assessment using polychaete Nereis virens for marine
samples and oligochaete Lumbriculus variegatus for freshwater samples.
MATERIALS AND METHODS
Chemicals
Calibration solutions were prepared from P AH standards (method standards for
wastewater, 16 compounds 0.1 mg mL- 1) (Chromatographie Specialties Inc., Brockville,
Canada). B700, HPCD and BuOH (99.4+%) were purchased from Sigma-Aldrich
(Oakville, Canada). Acetonitrile (HPLC grade) and DCM for liquid chromatography
analysis were purchased from VWR Ltd (Mississauga, Canada). The main physical and
chemical properties of extractants are given in Table 1.
30
Table 1. Sorne chemical and physical properties of the extractants
Formula MW Molecular Boiling Point
(g mor l) Volume (A3) (OC)
DCM CH2Ch 84.93 92 39.75
BuOH C4H90H 74.12 152 117.7
HPCD C!OSHI960S6 1460 262 Decompose
B700 ClsH37(OCH2CH2)nOH 4670 VariableA Decompose
(HLBB = 18.8) (n-100)
A Surfactants aggregate into supermolecular structure in water solution forming large size
micelles
B Hydrophilic-Lipophilic Balance
31
Sediment samples
Low contaminated marine sediments (total PAHs < 21-lg g - 1) were collected in the St.
Lawrence Estuary and Saguenay Fjord (Eastern Canada) in July 2004 with a Van Veen
grab. Marine sediments were obtained at low tide in the Kitimat Fjord (Western Canada) in
September 2003 in the vicinity of an operating aluminum smelter. Freshwater sediments
were collected in the St. Louis River (SLR) (Que bec, Canada) (one low contaminated and
two highly contaminated sites) in October 2003 with a hand corer also in the vicinity of an
operating aluminum plant. In aIl cases, whole sediment samples were stored in clean and
tight buckets and frozen at -20°C until analysis. A small portion (100 g) of each sample
was freeze-dried for 48 h, finely ground and stored in glass desiccators . Samples were
classified following their low or high content in total P AHs (LP AHs) and their main
geochemical properties are detailed in Table 2. The organic carbon content (Corg) was
determined by a carbon analyzer (Costech, Elemental Combustion System, CHNS-O, ECS
4010) after acidification of the sample with HCI (10 %) and digestion at 80°C for 15 h to
eliminate carbonates.
Table 2. Geochemical properties of the sediment samples
Samples % Humidity % Clay % Silt
Minette Bay 22.4 0 0
EGSL04-01 56.9 3.4 51.0
SLR Upstream 45.3 12.0 87.3
Hospital Beach 18.0 0 0
EGSL04-07 54.1 4.3 73.6
EGSL04-13 40.3 5.8 73.5
SLR Downstream 38.0 23.4 76.5
SLR Outflow 36.6 20.9 77.8
Scow Grid 53.8 1.9 48.1
Scrubber 44.5 6.5 93.5
B Lagoon 53.8 5.2 89.9
% Sand % Corg
100 0.04
45.6 1.2
0.7 5.1
100 0.08
22.1 1.1
20.7 0.7
0.1 2.5
1.3 2.0
50.0 2.41
0 41
4.9 29.9
L12PAHs
(flg gol w.w.) --0.06
0.07
0.3
0.3
0.4
1.1
25.5
279
811
4442
5723
V-.l N
33
DCM extraction
Each dry and powdered sediment sample (1.0 g) was extracted with 10 mL of DCM
by mechanical shaking for 16 h into a 35-mL Teflon tube. A surrogate solution (50 ~L) of
anthracene_d IO and fluoranthene_d IO at 0.4 ~g mL- I was added before shaking to determine
the recovery yield. The tube was centrifuged for 20 min at 3000 rpm and the supematant
transferred to a conical flask. The sample was concentrated under a nitrogen flow to 0.3-05
mL, diluted in pentane (2 mL), and concentrated again to 1.0 mL. The sample was eluted
through a clean-up minicolumn (Supelclean ENVI - 18 SPE 3 mL, Supelco) with 5 mL of
pentane:DCM (90: 10). The sample was concentrated in acetonitrile to 1.0 mL in an ice bath
to minimise loss of lighter P AHs. [25] Samples were then stored in a freezer at -6°C until
analysis. For comparison purposes, results were converted into wet weight (w. w.)
concentrations using the known percentage humidity of each sample.
BuOH extraction
Extractions were carried out in 35-mL Teflon tubes filled with 1.0 g of wet sediment,
10 mL of BuOH and 50 ~L of a surrogate solution of anthracene _ d IO and fluoranthene _ d IO
at 0.4 ~g mL - 1. The tubes were sealed and shaken overnight (16 h) at ambient temperature.
The tubes were then centrifuged for 20 min at 3000 rpm and the supernatant transferred to a
conical flask. The sample was concentrated under a nitrogen flow to 1.0 mL in an ice bath
to minimise loss of lighter PAHs. [15] The samples were then stored in a freezer at -6°C until
analysis.
34
HPCD extraction
Extractions were carried out in 35-mL Teflon tubes filled with 1.25 g of wet
sediment, 25mL of HPCD (50 mM) aqueous solution and 50 ilL of a surrogate solution of
anthracene_d lo and fluoranthene_d lO at 0.4 Ilg mL- I. The tubes were sealed and shaken
overnight (16 h) at room temperature. The tubes were then centrifuged for 20 min at 3000
rpm. A 0.5 mL aliquot of the supernatant was then withdrawn and diluted with
methanol:water (50:50) solution in a 10-mL volumetrie flask . The role of the methanol was
to decompose cyclodextrin complexes before analysis.[12,261 The samples were then stored
in a freezer at -6°C until analysis.
Surfactant extraction
Surfactant solutions were prepared using deionised water to reach a concentration of
lOg L - 1. Triplicates of analysed wet sediment (l.0 g) were weighted accurately into 35-mL
Teflon tubes, a surfactant aqueous solution (10 mL) and 50 ilL of a surrogate solution of
anthracene_d lo and fluoranthene_d lO at 0.4 Ilg mL- 1 were th en added to each tube. The
tubes were sealed and placed on a mechanical shaker for 16 h at ambient temperature. The
tubes were then centrifuged at 3000 rpm for 20 min. Each supernatant solution was
transferred into a separatory funnel, and 10 mL of DCM was added to transfer the extracted
PAHs into the organic phase. Water-soluble surfactants were not quantitatively extracted by
DCM and did not interfere with subsequent liquid chromatography analysis. The organic
phase was removed and the extraction with DCM was repeated once more. Organic
35
fractions were combined in a glass tube and the sample analysed as described above for the
DCM extraction. [27] The samples were then stored in a freezer at -6°C until analysis.
Optimization of B700 extraction
Preliminary work with surfactants indicated that their efficiency to extract P AHs from
sediments varied following their molecular structure and HLB (hydrophobic-lipophobic
balance). [27] Selection criteria for the choice of the best performing surfactant were based
on their efficiency to extract P AHs (results not shown) and the best reproducibility of the
extraction method. High-molecular-weight B700 turns out to be the most efficient
surfactant tested and the reproducibility of successive extractions of the same sample
reached 99.9 ± 1.5 % (Table 3).
The relationship between B700 concentrations in aqueous solution and the extraction
efficiency (Fig. 1) shows that the concentration of total P AHs extracted from contaminated
sediment increased with the increase of the surfactant concentration up to an optimum
value of ~5.25 x l0-3M and then reached a plateau. This phenomenon is known for
surfactant extraction and corresponds to the CMC, [28] the concentration above which
micelle formation becomes appreciable with a maximum hydrophobic effect and a
maximum retenti on capacity of neutral lipophilic molecules. [29,3 0] Based on these results, a
surfactant concentration that corresponded to 5.25 x l0-3 mol L- 1 was chosen for the B700
extraction.
36
Table 3. Reproducibility of the B700 extraction repeated three times on the same marine
sediment
PAHs concentration (~g g-I w. w.) Mean± SD
Fluorene N/DA N/DA N/DA N/DA
Phenanthrene 0.047 0.051 0.052 0.050 ± 0.003
Anthracene N/DA N/DA N/DA N/DA
Fluoranthene 0.216 0.218 0.213 0.216 ± 0.003
Pyrene 0.333 0.334 0.329 0.332 ± 0.003
Bz( a )Anthracene 0.049 0.046 0.042 0.046 ± 0.004
Chrysene 0.082 0.082 0.079 0.081 ± 0.002
Bz(b )F1uoranthene 0.025 0.022 0.021 0.023 ± 0.002
Bz(k)Fluoranthene 0.021 0.022 0.018 0.020 ± 0.002
Benzo( a )Pyrene 0.015 0.014 0.013 0.014 ± 0.001
Dibz( a,h)Anthracene 0.001 0.001 0.001 0.001 ± 0.000
Bz(g,h,i)Perylene 0.019 0.018 0.019 0.019 ± 0.001
LPAHs 0.808 0.808 0.787 0.801 ± 0.012
A Below detection limit
37
Fig. 1. Extraction efficiency of P AHs from marine sediment with a molarity gradient of
commercial surfactant B700. The optimum concentration has been determined at the
beginning of the plateau.
0.5
'01) 01) :::1. ~ 0.3 ~ o .-...... ro ~ 0.2 (]) u ~ o u ::r: 0.1 -< 0...
0.0 0.000
• 5.25 X 10-3 mol L- 1
• R2 = 0.98
0.002 0.004 0.006 0.008 Surfactant concentration (Brij ® 700) (mol L- 1)
38
Bioaccumulation studies
Polychaetes Nereis virens (3 to 5 g w.w.) were collected at low tide in Parc National
du Bic, near Rimouski along the St. Lawrence Estuary (Canada) . Organisms were gently
screened from the sediment, transported to the laboratory, and acc1imated in a 15- 20-cm
layer of their native sediment overlaid with flow through marine water for 10 days. The
worms were fed three times during these 10 days with commercial goldfish food. To
estimate PAH bioavailability in marine sediments, polychaetes were placed in one-litre
beakers (one polychaete per beaker) that contained 300 mL of sediment to be tested and
700 mL of seawater. Beakers were placed in an aquarium with flow-through seawater and
constant aeration. At each sampling time, several worms were removed from sediment,
rinsed in deionised water and allowed to purge their gut contents by standing for 8 h in
circulating seawater. [3 1]
Freshwater oligochaetes Lumbriculus variegatus, ~2 cm in length, were obtained
from Aquatic Research Organisms (Hampton, NH). As soon as the worms were received,
they were added to the St. Louis River sediments to start the experiment without a
depuration period because these worms originated from a rearing farm and were not
previously exposed to contaminated soil. Exposures of worms were conducted in 500-mL
glass beakers that contained 200 mL of sediment to be tested and 300mL of dechlorinated
tap water in a flow-through aquarium with aeration. The number of organisms per beaker
was 60, which provided a total mass of 0.4 to 0.5 g w.w. For each sampling, animaIs were
gently sieved from the sediment, rinsed in deionised water and allowed to purge their gut
for 6 h in c1ean dechlorinated tap water.[32] Polychaetes and oligochaetes were not fed
39
during the experiment. After the gut purging period, worms were put in scintillation vials,
freeze-dried at -20oe for 58 h, and dry residues crushed in a fine powder for chemical
extraction and analysis .
Worms (polychaetes and oligochaetes) were sampled seven times (at 0, 1, 3, 5, 7, 14
and 28 days) in triplicate (3 beakers). A total exposure period of 28 days was considered as
a sufficient time to ensure a steady state tissue concentration in both polychaetes and
0ligochaetesY3,34) As shown for four different samples (Table 4), the steady state plateau
was reached after 3 to 14 days. General conditions of the exposure system (pH, water
temperature, sali nit y for marine samples, dissolved oxygen and conductivity) were
monitored at each sampling day and found constant for the exposure period. Only results at
28 days are reported here.
40
Table 4. Examples of kinetic data for bioaccumulation of total P AHs (Ilg g -1 )
mean ± s.d., n = 3
Time (day) Bioaccumulation of total PAHs (Ilg g- ')
Minette Bay SLR Upstream Scrubber B Lagoon
0 0.07 ± 0.002 0.39 ± 0.01 2.11 ± 1.01 0.83 ± 0.03
1 1.19±0.01 0.52 ± 0.02 104.7 ± 28.8 7.3 ± 0.3
3 1.30 ± 0.01 0.92 ± 0.04 153 .5 ± 20.3 15 .9 ± 0.3
5 1.32 ± 0.02 1. 19 ± 0.01 163 .7 ± 10.9 20.8 ± 1.1
7 1.32 ± 0.01 1.22 ± 0.01 167.1 ± 11.6 21.5 ± 0.9
14 1.32 ± 0.01 1.24 ± 0.01 167.3 ± 9.4 21.5 ± 1.4
28 1.33 ± 0.01 1.24 ± 0.02 168.7 ± 12.2 22.4 ± 0.09
41
P AHs in biological tissues
Surrogate solution (50 ilL of anthracene_d lO and fluoranthene_d lO at 0.4 Ilg mL- 1)
was added to 200 mg of dried and ground worm tissue in a 35-mL Teflon tube and stored at
4°C for 16 h. Hexane:acetone (50:50, 5 mL) was then added to the mixture. The tube was
placed into an ultrasonic bath for 30 min. The tube was then mechanicalIy shaken for 3 h
and then returned again to the ultrasonic bath for 30 min. The tube was centrifuged for 15
min at 3000 rpm and the supernatant transferred to a conical flask. The extract was
concentrated under a nitrogen flow at ambient temperature to a volume of 0.5 mL. The
extract was eluted through a clean-up mini-column (Supelclean ENVI - 18 SPE 3 mL,
Supelco) with 5 mL of hexane:acetone (90: 10). The resulting clear solution was
concentrated to near dryness under nitrogen flow and re-dissolved in ~ ImL of acetonitrile
and evaporated again to 200 ilL in an ice bath to minimise loss of lighter P AHs. Analyses
were conducted on triplicate samples for each organism and sampling time.
P AHs analysis
AlI P AH analyses were carried out by liquid chromatography (LC) with fluorescence
detection. The apparatus consisted of a Rheodyne injector with a 20-IlL injection loop, a
Shimadzu LC-I0AD pump, a Supelcosil LC-PAH column (25 cmx 3 mm), and a Spectra
System FL3000 fluorescence detector. AlI analyses were done at a constant flow rate of 0.8
mL min - 1 with a mixture of nanopure water and acetonitrile as a mobile phase. The pump
program began at 75% acetonitrile and increased to 95% in 10 min and then to 100% in the
next 10 min, with a final plateau of 10 min. The cycle returned to 75% acetonitrile after a
42
total run time of 35 min. The excitation wavelength of the fluorescence detector was settled
at 280 nm and the emission was detected at 340 nm during the first 5.5 min for lighter
P AHs and th en shifted to 410 nm until the end of the pro gram for heavier P AHs.
Quantification was based on peak areas using five calibration solutions (0.002-005
/lgmL - 1). For quality control, a 0.5 /lgL - 1 PAHs Mix Standard Solution (Supelco) was
analysed every 10 samples. Fluorene (FLU), phenanthrene (PHE), anthracene (ANT) ,
fluoranthene (FLT), pyrene (PYR), benzo(a)anthracene (BaA), chrysene (CHR),
benzo(b )fluoranthene (BbF), benzo(k)fluoranthene (BkF), benzo(a)pyrene (BaP),
dibenzo(a,h)anthracene (DBA) and benzo(g,h,i)perylene (BPE) were quantified. Five
successive injections of the same extract provided a variability of < 1 0% for each identified
and quantified peak.
Two deuterated PAHs (anthracene_d lO and fluoranthene_d lO at 0.4 /lg mL- 1) were
used as internaI standards and were spiked in sediments and worm tissues before extraction.
Recovery of the internaI standards ranged from 76.3 ± 9.5 % (mean ± standard
deviation) to 98.7 ± 2.5 % (extractions of worms), from 88.9 ± 2.1 % to 91.2 ± 2.1 %
(extraction by DCM), from 89.6 ± 6.9 % to 90.5 ± 3.1 % (extraction by BuOH), from 85.8
± 2.0 % to 94.9 ± 6.4 % (extraction by B700), and from 88 .9 ± 2.8 % to 94.3 ± 4.3 %
(extraction by HPCD).
Data treatment
The results obtained by mild solid/liquid extractions and from worms allocated the
calculation of the extraction efficiency (% XTRAC) for each sediment. % XTRAC is
defined as the following ratio:
%XTRAC = Q worms xlOO Q DCM
43
where Qworms (Ilg g- l w.w.) and QDCM (Ilg g- l w.w.) stand for quantities extracted from
worms and by DCM, respectively.
RESULTS
St. Lawrence Estuary (EGSL04-01 and -07) and Saguenay Fjord (EGSL04-13)
sediments show quite similar grain-size distribution and percentage organic carbon (Table
2). Total PAHs in these deep-water marine sediments were aIl below 2 Ilg g- l (w.w.) with
dominance of 3-4-ring aromatics such as phenanthrene, fluoranthene, pyrene, chrysene and
benzo(b )fluoranthene (Fig. 2). Particulate organic matter present in these coastal sediments
is mainly derived from marine productivity and debris from terrestrial vascular plants. [35- 37]
Freshwater sediments from the St. Louis River (SLR samples) are also dominated by silt
with a relatively high proportion of clay. Sample SLR Outflow was collected near an
industrial outflow pipe from a nearby aluminum smelter and shows a relatively high level
of PAHs dominated by 4-5 rings such as pyrene, chrysene, bz(b)fluoranthene,
bz(k)fluoranthene and bz(a)pyrene (Fig. 2). Marine samples from Hospital Beach and
Minette Bay are clean sandy sediments taken at low tide and show very low P AH levels
and carbon content (Table 2). Two other samples were taken in the vicinity of an aluminum
plant (B Lagoon and Scow Grid) and contained high levels of aluminum smelter residues
mixed with natural muddy sediment and organic matter from a vegetated drainage basin.
44
FinaIly, one more sample was taken from the aluminum scrubber room after neutralisation.
The P AH pattern of these last three industrial samples exhibits high levels of aIl analysed
P AHs (Fig. 2).
45
Fig. 2. Distribution pattern of P AHs in four typical samples used to compare extraction
methods and bioaccumulation using worms.
1.00
1 0.80 0.60
,-.. 0.40 3 0.20 3
-;
: 0.16 20 c:
.~ 0.12 e c tU U c: o u
0.08
0.04
1000i 800 600
î 400
3 360 'Ol)
Ol)
20 c: o 'iii .... C tU U c: o u
320 280 240 200 160 120 80 40 o
/
EGSL04-13
l
Scrubber
l /
r1 n n
PAH
100
1 80 60 40
,-..
3 24 3
:.-
20
16
12
8
4
o
1200
1000 ,-..
3 3 800 -; Ol)
Ol) ~ '-" 600 c: .S: 'iii t: c: 400 tU U c: 0 u
200
0
SLR Outtlow
lin l ;,
,-,nn n B Lagoon
r-
n n
PAH
46
Results of extraction and bioaccumulation experiments are glven In Table 5. %
XTRAC of total PAHs is the yield of the chemical extraction or the worm bioaccumulation
when compared with the DCM extraction, which is considered the method of extracting aU
P AHs present in sediment samples. In sorne low contaminated samples, it was often
possible to extract more P AHs with BuOH and HPCD than with DCM. Similarly, worms
bioaccumulated more PAHs than extracted with DCM. Only surfactant B700 was less
efficient for extracting available P AHs in these samples. The efficiency of extractants was
strongly reduced in highly contaminated sediments although BuOH remained quite strong
with a yield that ranged from 27 to 58%. However, when comparing extractant
performance with PAHs bioaccumulated in worms, it becomes clear that HPCD and B700
ex tracts provided a much more realistic level of P AHs than BuOH, which grossly
overestimated the amount of total P AHs available to the worms in aIl cases.
Table 5. Results of extractions and bioaccumulation experiments with %XTRAC given in parentheses . Extractant values
followed by a star are significantly different from the corresponding worm bioaccumulation value. This is an indication of the
failure of the ex tractant to match the value of total PAHs observed in worms (P>O.Ol) (mean ± std. dev., n = 3)
Sediments DCM extraction BuOH extraction HPCD extraction B700 extraction Bioaccumulation (28 days)
(~g g-l w.w.) (~g g-l w.w.) (~g g-l w.w.) (~g g-l w.w.) (flg g-l w.w.)
Minette Bay 0.06 ± 0.02* 2.5 ± 0.5* 0.4* 0.02 ± 0.001 * 1.3±0.1 (> 100%) (> 100%) (33%) (>100%)
EGSL04-01 0.07 ± 0.03* 0.06 ± 0.01 * 0.3* 0.007 ± 0.004* 0.8 ± 0.3 (80%) (> 100%) (10%) (>100%)
SLR Upstream 0.3 ± 0.1 * 1.2 ± 0.4 1.4 0.09 ± 0.007* 1.3 ± 0.06 (> 100%) (> 100%) (26%) (>100%)
Hospital Beach 0.3 ± 0.05* 0.07 ± 0.004* 1.0 0.03 ± 0.009* 1.0 ± 0.2 (23%) (>100%) (10%) (>100%)
EGSL04-07 0.4 ± 0.3* 0.2 ± 0.03* 0.4* 0.011 ± 0.002* 0.6 ± 0.1 (50%) (>100%) (3%) (>100%)
EGSL04-13 1.1 ± 0.2* 0.5 ± 0.2 0.3* 0.04 ± 0.01 * 0.9 ± 0.1 (45%) (27%) (4%) (86%)
SLR Downstream 25 .5 ± 2.5* 11.9 ± 0.5* 0.2* 0.7 ± 0.2 0.8 ± 0.04 (46%) (0.9%) (2.7%) (3.3%)
SLR Outflow 279 ± 28* 163 ± 13* 4.7* 8.9 ± 4.1 8.9 ± 0.2 (58%) (1.7%) (3.2%) (3.2%)
Scow Grid 811 ± 154* 186 ± 36* 2.5* 1.6 ± 0.3 1.5 ± 0.2 (23%) (0.3%) (0.2%) (0.2%)
Scrubber 4442 ± 538* 1653 ± 105 26.9* 86 ± 8 86 ± 9 (36%) (0.6%) (1.9%) (1.9%)
B Lagoon 5723 ± 190* 1536 ± 107* 5.4* 30.7 ± 4.5 21.3 ± 9.4 (27%) (0.1 %) (0.5%) (0.4%)
-+:>. -...l
48
First, we examined the relative efficiency of chemical extractants to remove PARs
from sediment foUowing their molecular size, by classifying PARs by their number of
rings. For samples highly contaminated by aluminum smelter wastes it is remarkable to see
how BuOR and B700 work similarly to DCM for aU samples whereas the RPCD solution
seems to behave differently at least in the case of SLR Downstream and possibly with
Scow Grid samples (Fig. 3). Assuming DCM ex tracts nearly aU the PARs present in these
sediments, it becomes clear that non-exhaustive extractions with BuOR and B700 succeed
to mimic, in most cases, the exhaustive extraction when considering only a relative
proportion of ring numbers. The right column in each panel (Fig. 3) gives the relative
proportion of 3-, 4-, and 5- to 6-ring PARs found in worms after 28 days. In most cases the
proportion of 4-ring PARs is enhanced when compared with DCM and mild extractants,
and the proportion of 5- to 6-ring PARs is reduced. Again the RPCD solution does not
perform weil in the SLR Downstream and Scow Grid samples whereas B700 does very
weU except with ScowGrid. An early interpretation of these results would lead to the
conclusion that mild extractants provide a good means to predict bioaccumulation of PARs
by their molecular size in worms, but the extraction of low contaminated sediments tells us
another story.
100%
40% -
13 rings 4 rings
MM 5-6 rings
J 3 rings 4 rings 5-6 rings
DCM 0700 eubil 1-IFC0 Worms
100%
M 8700 80011 HPCD Worms
100%
80%
7011, -
49
Fig. 3. Comparison of extraction yields and worm bioaccumulation of PAH groups in
highly contaminated sediments. Amounts of PAHs are normalized to the sum of PAHs
extracted with each method.
B Lagoon
SLR OuttIow
nr.in R700 AuesH HPCS1 Worms
Extraction techniques and Moaccumu lation
Scrubber
M'$4 n'Inn FSC1H HPr71 Akwm
atraction techniques and boaccumulation
SLR Downstream
Extraction techniques and bioaccumulation
Scow Gnd
Extraction techniques and bioaccumulation
_ _1 3 rings 4 rings 5-6 rings
DCLA 8700 8u0H HPCO Worms
Extraction techniques and bioaccumulation
49
Fig. 3. Comparison of extraction yields and worm bioaccumulation of P AH groups in
highly contaminated sediments. Amounts of P AHs are normalized to the sum of P AHs
extracted with each method.
8 Lagoon
'" ~ Q.
EXll aCllOTllechnlQues arld bloaccumulation
Scrubber
OCI,I 8 7110 BuOH HPço Worms
E;dr action techniqu os and bioElccumu latioo
&;aw Gnd
ExtIaction techniQu es and bioaccumulaticn
100%
W'K,
W i,
4.0',)(,
111'110
SL Outrlow
~ 3 rings c::::J ~ rings _ 5.6 rlog$
Extracllon techniques and booaccumu lallon
SLR Downst ream
OC , Ell 00 BlDH HI'CO W<rm'
::=J 3 rings c::::J 4 rings _ 5-6 rln9s
Extraction techniques and bioaccumu lation
:=:J 3 rings c::::J 4 rinlls _ 5-6(10 9 5
50
We first observed that DCM extracted fairly constant proportions of 3, 4, 5 to 6 rings
In similar samples (EGSL04-01 , -07, and -13) and even in dissimilar samples from
freshwater (SLR Upstream) and west coast samples (Hospital Beach and Minette Bay)
where 3 rings account for from 20 to 40% and 5- 6 rings for from lOto 30%. HPCD and
B700 solutions provided inconsistent results when compared with DCM, whereas BuOH
was often quite close to DCM results except for Minette Bay (Fig. 4). Worms exposed to
low contaminated sediments showed a distribution pattern of bioaccumulated P AHs quite
the same in aH the studied stations (4 rings> 3 rings> 5-6 rings with proportions of 63- 87%,
11-34% and 2-7%, respectively). Very low amounts of 5-6 rings were accumulated in
worms although a large proportion of these heavy P AHs were consistently extracted by
solvents and extraction solutions. Neither DCM nor mi Id extractants correctly predicted the
proportion of 4-ring P AHs biaccumulated in both freshwater and seawater worms. AH
results showed a similar behaviour for both worm species and were pooled together for
further discussion.
51
Fig. 4. Comparison of extraction yields and worm bioaccumulation of PAH groups in low
contaminated sediments. Amounts of P AHs are normalized to the sum of P AHs extracted
with each method.
EG L04.{)1
Exlrachon tecMlques arod bioaccumu lalion
EGSL04.{)7
OCM àroo !luCH HPaJ INOOl1S
Extraclion I" chroiques arod bioaccumu lalion
EGSL04-13
OCM BroO BuOH HPCO IN om,.
E.1raction techniques and bioaccumu lation
HOSPltat Beach
nr. .. , R?nn RI ()H HP<":O IfJ .. ", ..
[:::=J 3 ri " os c::J 4 rings _ S·Orlogs
ExtracllQ. tecMlqull!$ and bioaccuml,l lalloo
Mmette Bay
oc ., B100 BuOH HPCO Worm.
'--1 3 rings r:::::J 4 ringS _ 5·6 rings
Extra cCi on teclmiques ilnd biotl ccumu lation
SLR Upstream
[)CM B700 BUOH HPCO W",m .
c::J 3 rings 4 rings
_ 5-6 110 g$
Extractioo techniques and bioaccumu lation
52
We also examined the relationship between individual P AHs bioaccumulated in
worms (both species) at a steady state and their bioavailablility as predicted by each
extractant for both groups of samples pooled together. Fig. 5 and Table 6 illustrate the
relationship between extracted and bioaccumulated low molecular-weight (LMW: fluorene,
phenanthrene, anthracene, fluoranthene, pyrene, benzo( a)anthracene, and chrysene) and
high-molecular-weight (HMW: benzo(b,k)fluoranthene, benzo( a )pyrene,
dibenzo( a,h)anthracene, and benzo(g,h,i) perylene) P AHs.
PAH lability as determined by BuOH (Fig. 5a) was not in good agreement with
worms data for both LMW and HMW P AH groups with high slopes and low correlation
coefficients (Table 6). In the opposite direction, P AH lability as determined by HPCD (Fig.
5b) was lower, particularly with LMW compounds, than bioavailability for worms although
discrepancy between both datasets is much lower than with BuOH. Finally, PAH lability, as
determined by B700 (Fig. 5c) was found in good agreement with worms data particularly
for LMW PAHs with a slope of 1.0 and R2=0.83. Data points for these compounds are
closely distributed around the dotted line, which represents al: 1 relationship. Predicted
PAH bioavailability was not in as good agreement for HMW PAHs (R2=0.74) (Table 6).
When LMW and HMW PAH data are included in linear regression analysis, the B700
method gave the best estimate of PAH bioavailability. Using zero-intercept fitting, linear
regression resulted in a slope of 1.01 and R2=0.82 for B700 compared with a too low slope
(0.23) for HPCD and a far too high si ope (18.46) for BuOH with poor R2 in both cases.
53
Fig. 5. Low CLMW) and high CHMW) molecular weight PAHs extracted by BuOH Ca),
HPCD Cb) and B700 Cc) as a function ofbioaccumulated PAHs in worms over 28 days. The
dotted lines represent a hypothetical 1: 1 correlation.
Ca) LMNPAHs HMN PAHs
450 350 lL 025 L ' 400 . " • 05 .' 300 350
. ,., / / . o . • • 250 0
~ 300 0 0.5 1 ~ 0 025 ~ ~
-Dl 250 -Dl 200 Dl Dl .3 200 • .3 150 I • • I 0 150 0 :J :J 100 en 100 en
50 50 •
0 ~ .... - - --0 ---------
0 5 10 15 20 0 1 2 3 4 Worrrs (jJg g" w.w.) Worrrs (jJg g" W .w .)
Cb) LMN PAHs HMNPAHs
20 5
'b '~ • 0: /. / ~/ •• 4 0.5 h
15 /
~ ~ o ~ ~ 0 0.5 1 ~ 3
"Cl "Cl .3 10 / Dl .3
0 0 2 0
"," "'. 0 Cl.. Cl.. I 5 / • I
0 0 0 5 10 15 20 0 2 3 4
Worrrs (jJg g" w.w.) Worrrs (jJg g" W .w .)
CC) LMW PAHs HMWPAHs
25 10
05~ o~Lt • r· / / 20 8 . / / • /.
o .-.-.., 0 ~ ~ 0 0.5 ~ 15 0 0.5 1 ~ 6 bl
., • Dl
.3 10 .3 4 0 • 8 0 ,.... ,.... en en
5 2
• • 0 0 0 5 10 15 20 0 1 2 3 4
Worrrs (jJg g" w.w.) Worrrs (jJg g" w .w .)
Table 6. Linear correlation parameters obtained when comparing LMW and HMW P AHs extracted with BUOH, HPCD and
B700 with P AHs bioaccumulated in worms for both low and high contaminated sediments
The first line for each P AH group gives results for the intercept forced to zero and the second is for free intercept.
PAHs
LMW
HMW
AU PAHs
Bu OH extraction
Slope
16.98
16.16
93 .71
93.92
18.46
17.28
Intercept R2
0.0 0.48
8.98
0.0
-0.45
0.0
Il.90
0.49
0.95
0.95
0.42
0.44
HPCD extraction
Slope
0.22
0.21
0.91
0.87
0.23
0.22
Intercept R2
0.0 0.57
0.09
0.0
0.07
0.0
0.14
0.58
0.66
0.67
0.48
0.50
B700 extraction
Slope
1.00
1.01
1.65
1.65
1.01
1.01
Intercept R2
0.0 0.83
-0.06
0.0
0.006
0.0
0.02
0.83
0.74
0.74
0.82
0.82
VI +:>.
55
To gam a better understanding on the capacity of the B700 method to predict
individual P AH bioaccumulation in worms, the ratio of P AH concentrations determined by
B700 extraction to PAH concentrations found in worms were plotted against the PAH
octanol-water partitioning coefficient (Kow) (Fig. 6). A ratio approaching 1 indicates the
ability to predict PAH bioavailability based on B700 extraction, whereas a ratio < 1
indicates a low chemical lability, and a ratio > 1 indicates an over-predicted PAH
bioavailability. Low contaminated sediments show too low ratios even for most high Kow
P AHs. However, highly contaminated sediments present a quite different picture as aIl high
log Kow (> 5.8) PAHs exhibit a ratio> 1 with high variability between replicates whereas
B700 extraction is a good predictor of PAH bioavailability for compounds with log Kow
values <5.8 (Fig. 6).
56
Fig. 6. Relationship between PAR octanol-water partitioning coefficient (Kow) and the
ability to predict PAR bioavailability using B700 availability methodology.
4
3
2
... .. · -ANT-·
FLU
Low contaminated sediments
DBA
BPE
FLT BaA PYR o ~--+---~~----~--~~~~~~----~-+~--~~--+----------,
4
- 1
8
7
6
5
4 PHE
3
2
FLU
5 6 7
Highly contaminated sediments
CHR
BaA
BaP
DBA
BkF BbF
BPE
L ...... .
8
o ~----~±+------~------~~--~------~~----~-------------, 4 5 6 7 8
- 1
57
DISCUSSION
In this study three mild extraction methods were used with marine and freshwater
sediments in an attempt to estimate the bioavailability of P AHs for invertebrates in constant
contact with sediments. Our results highlight the importance of considering both the level
of P AHs in sediments and the molecular size of P AHs when attempting to predict
bioaccumulation in a biological sampler like worms using a solid/liquid extraction method.
A surfactant B700 solution was quite successful to predict the PAH content in worms when
exposed to highly contaminated sediments and smelter residues, whereas BuOH extracted
lOto 50 times too much P AHs being a far too efficient solvent for absorbed P AHs. When
low contaminated sediments are use d, HPCD and BuOH were revealed to be better for
estimating bioaccumulation in worms whereas B700 appeared to be a too mild extractant
with a very low yield for most samples.
Extractable P AHs and sediment properties
The over 100% yield (% XTRAC) often observed with low contaminated sediments
using BuOH and HPCD might be in part an artifact introduced by the extraction method
using DCM. As DCM is water insoluble and do es not work properly with wet sediments,
samples have been freeze-dried before extraction whereas wet samples were used with mild
extractants. The drying process may have modified the structure of the organic matter and
reduced the availability of small amounts of P AHs to DCM extraction. These anomalous
results might also sirnply reflect analytical variability because of inhornogeneity of sorne
58
sub-samples for the same station. Extractant B700 did not give a yield of over 100% with
the lowest contaminated samples which possibly indicates its low ability to disaggregate
natural organic particles (pellets of invertebrates) usually present in natural sediments. Even
if Hospital Beach and Minette Bay have the same sand percentage (100%) and are both low
contaminated sediments, their % XTRAC by B700 was quite different (10 and 33%,
respectively). This difference might be related to the size of the sand particles and the
nature of the organic matter. Indeed, soils or sediments that are dominated by bigger
particles generally have a greater mineralisation of P AHs. [38] A closer examination of
particle size analysis (not shown) confirmed that Hospital Beach had finer sand particles
than the Minette Bay sample. BuOH also extracted much more P AHs from Minette Bay
than from Hospital Bay although DCM found less PAHs in Minette Bay.
The % XTRAC for two of the three St. Lawrence stations by B700 are quite similar
(3% and 4% for EGSL04-07 and EGSL04-13 , respectively) even though the third one is
higher (l 0%). These results imply a possible better sequestration for stations -07 and -13
than for EGSL04-01 , which has a higher sand percentage (45.6% compared with 22.1 and
20.7%) with a comparable level of organic carbon content. BuOH extraction shows the
same trend. Carmichael and Pfaender[38] showed that the silt and clay fractions of the soils
they studied were significantly linked to the P AH mineralised percentage. A quick
examination of Fig. 4 shows the disparity of results between chemical extractants when
looking at the molecular size of PAHs, particularly for St. Lawrence Estuary samples.
HPCD and B700 extracted relatively more 3- and 4-ring PAHs because of their better
solubility in water than 5-6-ring P AHs, which were better extracted by DCM and BuOH,
59
both being organic solvents able to dissolve organic lipids and reach sequestrated P AHs.
Worms did not extract 5-6-ring PAHs and simply retained 3- and 4-ring PAHs.
With highly contaminated sediments, BuOH extractions never exceeded the DCM
extraction efficiency, although BuOH appeared very efficient when compared with HPCD
and B700. Attempts to relate % XTRAC to the geochemical properties of highly
contaminated sediments were unsuccessful. The presence of huge quantities of P AHs
mainly from aluminum smelter residues overwhelms other physical or chemical factors in
the extraction process. The frequent observation of non-linear sorption isotherms for
organic compounds introduced into soil suggests a saturation of sorption sites. [39] Such a
saturation would be reflected in a declining percentage of the chemical that is sequestered
as the concentration increases as shown by Chung and Alexander. [40] When examining the
pattern distribution of 3-, 4- and 5-6-ring PAHs for aIl chemical extractions in aIl heavily
contaminated sediments, it turns out that aU patterns are quite similar (except for
unexplained HP CD in the SLR Downstream sample),which confirms a similar behaviour of
these extractions where solubility of the adsorbed P AHs on inert particles becomes the
main factor that controls the efficiency of the extraction. As already observed for low
contaminated sediments, both worm species did not extract and bioaccumulate 5-6 rings
although a very large proportion of these heavy P AHs were present in these samples (Fig.
2).
In a recent paper, our laboratory examined the sorption and desorption properties of
some of the samples studied here using different chemical tools. [41] We observed that P AHs
in a highly contaminated B Lagoon sample were much less available to a desorption
60
process that P AHs in freshly released scrubber residues, although both samples showed a
quite similar distribution of extractable P AHs. This reduced availability was attributed to a
weathering pro cess that resulted in a stronger sequestration of P AHs in the 20-year-old
lagoon sediment. [41] This finding is supported by present results (Table 5) where % XTRAC
is higher with scrubber than B Lagoon samples for all chemical soft extractions and also
with worms. Worms bioaccurnulated a higher proportion of 3- and 4-ring PAHs in the
scrubber than in the B Lagoon sample (Fig. 3), which is in agreement with the hypothesis
that LMW PAHs in the scrubber were more available than in the lagoon sediment.
Nam and Alexander[42] suggested that the bioavailability of hydrophobic compounds
(such as PAHs) can be extensively reduced by particles that bear nanopores with
hydrophobic surfaces. Hydrophobic compounds may become sequestered and less available
to living organisms if they penetrate such porous materials. [43,44] Mayer[45] proposed that
organic matter in marine sediments is protected by its location inside pores too small to
allow the entrance or function of hydrolytic enzymes. Indeed, pores with a diameter of less
than 100 nm have been observed in a variety of dissimilar soils. [46,47] A molecule that is
entrapped in a nanopore with a diameter smaller th an 100 nm is probably unavailable to
any living organism since the smallest bacteria have a larger diameter. [42] Similarly, large
HPCD molecules and B700 micelles most probably cannot reach these nanopores, which
reduce their capacity to capture small P AHs. On the other hand, DCM and BuOH are much
smaller molecules (Table 1) and can more easily access the smallest pores, which means a
better extraction capacity. The situation appears different for 5-6-ring PAHs with their
large volume (229A3 for BaP) which prevents them from penetrating nanopores and leaves
61
only larger pores available for large HPCD and surfactant micelles to access. This might be
the reason why HPCD succeeded to predict the availability of high-molecular-weight P AHs
in worms (Fig. 5).
Chemically availability v. bioaccumulation
In this study soft extraction methods were used to estimate the bioavailable fraction
of P AHs from unspiked and untreated sediments. Several researchers have suggested that
mild BuOH extraction may be an appropriate means for predicting P AH
bioavailability. [9,10,48] However, our results indicate that BuOH-extractable PAHs do not
correlate with P AHs found in worms exposed to sediments for 28 days, as BuOH
overestimated by up to 540 times the P AH availability in highly contaminated samples. In a
recent study by Juhasz et al. using creosote contaminated SOil,[49l BuOH extraction
underestimated the 3-ring and overestimated the 4-, 5- and 6-ring PAHs compared with
biodegradation. Reid et al. and Macleod and Semple/ l3,50] using pyrene spiked soils, also
found an overestimation of pyrene bioavailability by BuOH extraction compared with
bacterial mineralisation. Conversely, Kelsey et al. and Liste and Alexander[9, 10] found good
relationships between P AH desorption using BuOH and P AH bioavailability (estimated by
microbial degradation and earthworm uptake assays) in laboratory spiked soil that
contained single P AH compounds. The above studies exemplify conflicting results present
in the literature on bioavailability research using spiked soils and low molecular- weight
primary alcohols. Spiking and aging soil or sediments under laboratory conditions for
weeks or months may not be long enough to truly reflect field conditions. In addition,
contaminant bioavailability may be influenced by the presence of other organic
62
contaminants; a situation not taken into account in single P AH spiked soil studies. As a
result, spiked soil studies (determining contaminant availability and sequestration) often
fail to mimic conditions found in field contaminated soils.[51]
In the present study, the assessment of PAH bioavailability using HPCD extraction
resulted in a good prediction of total P AH bioavailability in low contaminated sediment
samples, but relative proportions of 3-, 4-, and 5-6-ring P AHs did not reflect the
proportions found in worms. HP CD underestimated total P AHs in most highly
contaminated samples. HPCD was developed as an extractant for assessing contaminant
bioavailability because HPCD has a high solubility, the prevalence of hydroxy functional
groups on the exterior of the torus, and a hydrophobie organic cavity, makes it possible to
form an inclusion complex with PAHs.[13] HPCD extraction was in good agreement with
bioavailability as determined by mineralisation in phenanthrene spiked soil.[13] Conversely,
Cuypers et al. and Juhasz et al. [14,49] used petroleum-contaminated harbour sediments and
creosote-contaminated soil and found HPCD extraction predicted correctly 3- and 4-ring
P AH biodegradability, whereas the biodegradability of 5- and 6-ring P AHs was
overestimated.
This study presents a first attempt to assess PAH bioavailability with surfactant B700.
When aIl PAH data are included in the same linear regression analysis, the B700 method
gave the best estimate of PAH bioavailability (R2=O.82 and slope=l.Ol) (Table 6).
However, B700 is less efficient with low contaminated sediments, most probably because
of its limited capacity to penetrate lipidic particles or tissues where P AHs have been
accumulated by micro- and macro-invertebrates. In both high and low contaminated
63
sediments, worms showed a low proportion of 3-ring P AHs and a relatively high proportion
of 4-ring P AHs. This result seems to be linked to the high solubility and biodegradability of
3 rings, particularly phenanthrene. As worms were sampled after a continuous exposure of
28 days, light P AHs accumulated at the beginning of the exposure have been partly
metabolised and eliminated by worms whereas heavier 4-, 5- and 6-ring PAHs were much
less biodegraded. The result is an apparent low bioaccumulation of lower P AHs and an
apparent excess of higher P AHs when compared with P AHs in DCM extracts.
The contrast between low and highly contaminated samples is best illustrated ln
Fig. 6 where the ability of B700 to predict PAH bioavailablility is assessed following its
affinity for lipidic fractions (log Kow). Almost all PAHs exhibit a ratio < 1 in low
contaminated samples, which indicates that worms cannot only adsorb surficial P AHs but
also digest marine organic mater and extract P AHs already bioaccumulated by living and
dead organisms present in the sediment. In contrast, highly contaminated sediments showed
an interesting two-plateau pattern already observed by Juhasz et al. [49) when examining
predictor capability of HPCD for P AH availability in soil highly contaminated with
creosote- treated in biopile for 16 weeks. The fact that both HPCD and B700 still
overestimate the availability of P AHs with log Kow > 5.5 might be related to their molecular
size, which is smaller than enzymes and allows their accessibility to small interstices where
enzymes cannot enter.
64
CONCLUSION
Strong differences were observed in quantities and proportions of P AHs obtained
using different techniques for assessing the bioavailability of P AHs in sediments. Although
several researchers have suggested that mild BuOH extractions may be an appropriate mean
for predicting PAH bioavailability, our results disagree with previous studies for both low
and highly contaminated sediments where BuOH was found to extract much more P AHs
than worms can bioaccumulate in their tissues, and where HPCD underestimates the
bioavailability of P AHs especially for LMW P AHs. On the other hand, use of the surfactant
B700 revealed a good prediction for PAHs in highly contaminated sediments. Laboratory
studies conducted with freshly spiked or slurried sediment may overestimate the
bioavailability of contaminants compared with field situations where longer equilibration
times between sediment and persistent contaminants reduce the bioavailability.[52,531
Moreover, most of these studies determined the bioavailability by measuring microbial
degradation in laboratory reactors, an approach that might not be comparable with P AHs
bioaccumulated by worms as worms are not passive samplers and can modify the P AH
profile in their tissues.
Since risk assessments from contaminated soils are currently being overestimated by
chemical analyses that rely on an initial vigorous solvent extraction, and because
remediation may be suitable when none or a less vigorous treatment is required, it is
essential to include bioavailability assessment in regulatory decisions. Using worm uptake
and biodegradation are still the best tools to estimate P AH bioavailability, but biological
65
methods are time consuming and expensive. Our results illustrate difficulties in finding an
adequate chemical predictor of PAH bioavailabilty, particularly because PAH
concentrations and the sequestration process play a determining role in the quality of
results. Because B700 lS not expensive and solutions easy to prepare, an extraction
procedure involving this high-molecular-weight surfactant is revealed to be useful
particularly with sediments that show PAH levels > 21lg g- I and log Kow <5.8.
ACKNOWLEDGEMENTS
This study was supported by a grant from the Canadian Research Chair in Molecular
Ecotoxicology (E.P.) and from the Natural Science and Engineering Research Council of
Canada (NSERC) - Collaborative Research Development Pro gram (CRD). Samples have
been provided by Alcan Ltée. This paper is a contribution of the Quebec-Ocean Network.
[1]
[2]
REFERENCES
1. V. Pothuluri, C. E. Cemiglia, in Bioremediation: Princip/es and Practice (Eds S.
K. Sikdar, R. L. Irvine) 1998, vol. 2, pp. 462-520 (Publishing Company: Lancaster)
M. H. Van Agteren, S. Keuning, D. B. Janssen, Handbook on Biodegradation and
Biological Treatment of Hazardous Organic Compounds 1998 (Kluwer Academie
Publishers: Dordrecht)
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
66
J. J. Pignatello, in Reactions and Movement of Organie Chemieals in Soifs (Eds B. L.
Sawhney, K. Brown) 1989, pp. 271-304 (Soil Science Society of America, American
Society of Agronomy: Madison, WI)
P. B. Hatzinger, M. Alexander, Effect of ageing chemicals in soil upon their
biodegradability and extractability, Environ. Sei. Teehnol. 1995,29,537
B. J. Reid, K. C. Jones, K. T. Semple, Bioavailability of persistent organic pollutants
in soils and sediments - a perspective on mechanisms, consequences and assessment,
Environ. Pollut. 2000, 108, 103
J. D. MacRae, K. J. Hall, Comparison of methods used to determine the availability
of polycyclic aromatic hydrocarbons in marine sediments, Environ. Sei. Teehnol.
1998,32,3809
P. C. M. van Noort, G. Comelissen, T. E. M. ten Hulscher, B. A. Vrind, H. Rigterink,
A. Belfroid, Slow and very slow desorption of organic compounds from sediment:
influence of sorbate planarity, Water Researeh 2003, 37, 2317
C. Cuypers, T. Grotenhuis, J. Joziase, W. Rulkens, Rapid persulfate oxidation
predicts PAH bioavailability in soils and sediments, Environ Sei Teehnol, 2000, 34,
2057
J. W. Kelsey, B. D. Kottler, M. Alexander, Selective chemical extractants to predict
bioavailability of soil-aged organic chemicals, Environ Sei Teehnol, 1997, 31, 214
H. H. Liste, M. Alexander, Butanol extraction to predict bioavailability of P AHs in
soil, Chemosphere, 2002, 46, 1011
67
[II] V. Librando, O. Hutzinger, G. Tringali, M. Aresta, Supercritical fluid extraction of
polycyclic aromatic hydrocarbons from marine sediments and soil samples,
Chemosphere, 2004, 54, 1189
[12] M.-C. Chang, C.-R. Huang, H.-Y. Shu, Effects of surfactants on extraction of
phenanthrene in spiked sand, Chemosphere, 2000, 41, 1295
[1 3] B. J. Reid, 1. D. Stokes, K. C. Jones, K. T. Semple, Nonexhaustive cyclodextrin-
based extraction technique for the evaluation of PAH bioavailability, Environ. Sei.
Teehnol. 2000, 34, 3174
[14] C. Cuypers, T. Pancras, T. Grotenhuis, W. Rulkens, The estimation of PAH
bioavailability in contaminated sediment using hydroxypropyl-~-cyclodextrin and
triton X-100 extraction techniques, Chemosphere, 2002, 46, 1235
[15] 1. Tang, M. Alexander, Mild extractability and bioavailability of polycyclic aromatic
hydrocarbons in soil, Environ. Toxieol. Chem. 1999, 18, 2711
[1 6] 1. Tang, H. H. Liste, M. Alexander, Chemical assays of availability to earthworms of
polycyclic aromatic hydrocarbons in soil, Chemosphere, 2002, 48, 35
[1 7] W. 1. Shieh, A. R. Hedges, Properties and applications of cyclodextrins, J
Maeromol. Sc. Pure Part A, 1996, 33, 673
[1 8] G. Shixiang, W. Liansheng, H. Quingguo, H. Sukui, Solubilization of polycyclic
aromatic hydrocarbons by ~-cyclodextrin and carboxymethyl-~-cyclodextrin,
Chemosphere , 1998,37,1299
[1 9] X. Wang, M. L. Brusseau, Cyclopentanol-enhanced solubilization of polycyclic
aromatic hydrocarbons by cyclodextrins, Environ. Sei. Teehnol. 1995,29, 2346
68
[20) B. J. Reid, K. T. Semple, K. C. Jones, in Proc. Fifth International In Situ and On-Site
Bioremediation Symposium (Eds A. Leeson, B. C. Alleman) 1999, pp. 253-258
(Battelle Press: Columbus)
[21) A. L. Swindell, B. J. Reid, Comparison of selected non-exhaustive extraction
techniques to assess PAH availability in dissimilar soils, Chemosphere, 2006, 62,
1126
[22] F. Volkering, 1. 1. Quist, A. F. M. Van Velsen, P. H. G. Thomassen, M. Olijve, in
Contaminated Soil '98 1998, vol. 1, pp. 251-259 (Thomas Telford: London)
[23) 1. M. Banat, Biosurfactants production and possible uses in microbial enhanced oil
recovery and oil pollution remediation: a review, Bioresour. Technol. 1995, 5 J, 1
[24] F. Volkering, A. M. Breure, W. H. Rulkens, Microbiological aspects of surfactant use
for biological soil remediation, Biodegradation 1997, 8, 401
[25) D. Brion, E. Pelletier, Modelling PAHs adsorption and sequestration in freshwater
and marine sediments, Chemosphere, 2005,61,867
[26) X. Wang, M. L. Brusseau Solubilization of some low-polarity organic compounds by
hydroxypropyl-p-cyclodextrin, Environ. Sei. Technol. 1993, 27, 2821
[27] M. Barthe, Study of the Chemical Sequestration of Polycyclic Aromatic
Hydrocarbons (P AHs) by Selective Extraction of Freshwater and Marine Sediments
2002, M. Sc. Thesis (Université du Québec à Rimouski: Rimouski QC)
[28) S. Laha, Z. Liu, D. A. Edwards, R. G. Luthy, in Aquatic Chemistry: Interfacial and
Interspecies Pro cesses (Ed 1. 1. Morgan) 1995, pp. 339-361 (American Chemical
Society: Washington, DC)
69
[29] D. H. Everett, Basic Principles of Colloid Science 1992 (Royal Society of Chemistry:
London)
[30] P. C. Hiemenz, R. Rajagopalan, Principles of Colloid and Surface Chemistry 1997
(Marcel Dekker, Inc.: New York, NY)
[3 1] M. Christensen, O. Andersen, G. T. Banta, Metabolism of pyrene by the polychaetes
Nereis diversicolor and Arenicola marina, Aquat. Toxicol. 2002, 58, 15
[32] D. R. Mount, T. D. Dawson, L. P. Burkland, Implications of gut purging for tissue
residues determined in bioaccumulation testing of sediment with Lumbriculus
variegatus, Environ. Toxicol. Chem. 1999, 18, 1244
[33] H. Lee II, B. Boese, 1. Pelletier, M. Winsor, D. Specht, R. Randall, Guidance
Manual: Bedded Sediment Bioaccumulation Tests 1989, EPA/600/x-89/302 (U.S.
Environmental Protection Agency: Duluth, MN and Washington, De)
[34] U.S. Environmental Protection Agency, Methods for Measuring the Toxicity and
Bioaccumulation of Sediment-associated Contaminants with Freshwater
Invertebrates 2000, 2nd edition, EPA 600/R-99/064 (U.S. Environmental Protection
Agency: Duluth, MN and Washington, DC)
[35] 1.-C. Colombo, N. Silverberg, 1. N. Gearing, Biogeochemistry of organic matter in
the Laurentian Trough, I. Composition and vertical fluxes of rapidly settling
particles, Mar. Chem. 1996,51,277
[36] J.-c. Colombo, N. Silverberg, 1. N. Gearing, Biogeochemistry of organic matter in
the Laurentian Trough, II. Bulk composition of the sediments and relative reactivity
of major components during early diagenesis, Mar. Chem. 1996,51 , 295
[37]
[38]
[39]
70
P. Louchouarn, M. Lucotte, N. Farella, Historical and geographical variations of
sources and transport of terrigenous orgamc matter within a large-scale coastal
environment, Org. Geoehem. 1999, 30, 675
L. M. Carmichael, F. K. Pfaender, Polynuc1ear aromatic hydrocarbon metabolism in
soils: relationship to soil characteristics and preexposure, Environ. Toxieol. Chem.
1997, 16, 666
M. B. McBride, Environmental Chemistry of Soils 1994 (Oxford University Press :
New York, NY)
[40] N. Chung, M. Alexander, Effect of concentration on sequestration and bioavailability
oftwo polycyclic aromatic hydrocarbons, Environ. Sei. Teehnol. 1999, 33,3605
[4 1]
[42]
[43]
[44]
[45]
G. Breedveld, E. Pelletier, R. St. Louis, G. Cornelissen, Effect of concentration on
sequestration and bioavailability of two polycyc1ic aromatic hydrocarbons, Environ.
Sei. Teehnol. 2007, 41 , 2542
K. Nam, M. Alexander, Role of nanoporosity and hydrophobicity in sequestration
and bioavailability: tests with model solids, Environ. Sei. Teehnol. 1998, 32, 71
W. P. BaU, P. V. Roberts, Long-term sorption of halogenated organic chemicals by
aquifer materials: 2. Intrapartic1e diffusion, Environ. Sei. Teehnol. 1991, 25, 1237
S. L. Scribner, T. R. Benzing, S. Sun, S. A. Boyd, Desorption and bioavailability of
aged simazine residues in soil from a continuous corn field, J Environ. Qual. 1992,
21 , 115
L. M. Mayer, Surface area control of organic carbon accumulation in continental
shelfsediments, Geoehim. Cosmoehim. Acta 1994, 58, 1271
[46)
[47)
[48)
[49)
[50)
[51)
[52)
71
1. D. Sills, L. A. G. Aylmor, 1. P. Quirk, Relationship between pore distributions and
physical properties of clay soils, Aust. J Soil Res. 1974, 12, 107
J. Hassink, L. A. Bouwman, K. B. Zwart, L. Burssaard, Relationship between
habitable pore space, soil biota and mineralization rates in grassland soils, Soil Biol.
Biochem. 1993, 25,47
G. D. Breedveld, D. A. Karlsen, Estimating the availability of polycYclic aromatic
hydrocarbons for bioremediation of creosote contaminated soils, Appl. Microbiol.
Biotechnol. 2000, 54, 255
A. L. Juhasz, N. Waller, R. Stewart, Predicting the efficiency of polycyclic aromatic
hydrocarbon bioremediation in creosote-contaminated soil using bioavailability
assays, Bioremed. J 200S, 9, 99
C. J. A. Macleod, K. T. Semple, Sequential extraction of low concentrations of
pyrene and formation of non-extractable residues in sterile and non-sterile soil, Soil
Biol. Biochem. 2003,35, 1443
S. B. Hawthorn, D. G. Poppendieck, C. B. Grabanski, R. C. Loehr, Comparing PAH
bioavailability from manufactured gas plant soils and sediments with chemical and
biological tests. 1. P AH release during water desorption and super critical carbon
dioxide extraction, Environ. Sei. Technol. 2002,36, 4795
A. U. Comad, A. D. Comber, K. Simkiss, Pyrene bioavailability; effect of sediment-
chemical contact time on routes of uptake in an oligochaete worm, Chemosphere,
2002,49, 447
[53 ]
72
M. T. Leppanen, J. V. K. Kukkonen, Effect of sediment-chemical contact time on
availability of sediment-associated pyrene and benzo[a]pyrene to oligochaete worms
and semi-permeable membrane devices, Aquat. Toxicol. 2000, 49, 227
CHAPITRE II
PASSIVE SAMPLERS VERSUS SURFACTANT EXTRACTION FOR
THE EVALUATION OF P AH A V AILABILITY IN SEDIMENTS WITH
VARIABLE LEVELS OF CONTAMINATION
M. Barthe, É. Pelletier, G.D. Breedveld, G. Cornelissen
(Chemosphere, 2008, 71: 1486-1493)
74
Résumé
L'objet de cette étude était de tester l'efficacité des échantillonneurs solides passifs,
des bandes de polyoxyméthylène (POM) et des tubes de silicone polydiméthylsiloxane
(PDMS), à prédire le biodisponibilité des HAP présents dans les sédiments contaminés. Les
résultats ont été comparés aux données de bioaccumulation et d' une extraction
solide/liquide utilisant le tensioactif Brij® 700 (B700). Les deux échantillonneurs passifs
ont été trouvés à agir différemment. Le PDMS surestime la disponibilité des HAP dans tous
les sédiments étudiés. La méthode avec le POM produit des résultats en accord avec ceux
obtenus avec l' extraction au B700. Quoi qu' il en soit, le POM et le B700 sousestiment la
disponibilité des HAP dans les sédiments faiblement contaminés où les facteurs biologiques
(matière organique digestible) deviennent importants. La biodisponibilité des HAP totaux a
été prédite correctement par le POM et le B700 dans les sédiments contaminés par des
alumineries. Un examen plus précis des résultats des HAP individuels indiquait que les
deux techniques surestimaient la disponibilité des grosses molécules avec un log Kow >6
suggérant un mécanisme biologique limitant l' incorporation des plus gros HAP ce qui
semble être relié à la taille moléculaire des composés.
Mots clés: Échantillonneur passif, Extraction avec un tensioactif, Biodisponibilité,
Bioaccumulation par des vers, HAP, Sédiment.
75
Abstract
The purpose of this study was to test the efficiency of passive solid samplers,
polyoxymethylene (POM) strips and polydimethylsiloxane (PDMS) silicon tubing, to
predict the bioavailability of native P AHs in contaminated sediments. Results were
compared with worm bioaccumulation data and solid/liquid extraction using the surfactant
Brij® 700 (B700). The two passive samplers were found to act differently. The PDMS
sampler overestimated the availability of P AHs in aIl studied sediments. The POM method
provided results in accordance with those obtained with the B700 extraction. However,
POM and B700 methods underestimated PAH availability in low contaminated sediments
where biological factors (digestible organic matter) become important. Bioavailability of
total PAHs was correctly predicted by POM and B700 in highly contaminated aluminum
smelter sediments. A closer examination of individual P AH results indicated that both
techniques overestimated the availability of large molecules with log Kow >6 suggesting a
biological mechanism limiting uptake of larger P AHs which seems to be related to the
molecular size of compounds.
Keywords: Passive sampler; Surfactant extraction; Bioavailability; Worm
bioaccumulation; P AH; Sediment.
76
INTRODUCTION
Hydrophobie organic chemicals such as polycyclic aromatic hydrocarbons (P AHs)
are common contaminants in industrial soils and marine sediments (Kennish, 1997). Risk
assessment procedures of hydrophobie chemicals in soils and sediments are usually based
on total soil/sediment concentrations or pore water concentrations estimated from field
concentrations and generic organic carbon normalized partition coefficients (Doucette,
2003). These coefficients are generally based on experiments with freshly spiked standard
soils and sediments and rarely consider the strong sorption to carbonaceous geosorbents
(e.g. , soot, coal, kerogen) that may result in increases of sorption coefficients by 1-2 orders
of magnitude (Luthy et al., 1997; Cornelissen and Gustafsson, 2004).
Various chemical techniques have been developed to study bioavailability in soil and
sediment. Sorne approaches focussed on the extraction of the weakly bound fraction. Solid
sorbents such as Tenax (Yeom et al., 1996; Cornelissen et al. , 1997) or XAD-2 (Carroll et
al., 1994), solvents (Kelsey et al. , 1997), aqueous solutions with cyclodextrin (Cuypers et
al. , 2002; Swindell and Reid, 2006), surfactants (Chang et al., 2000; Cuypers et al. , 2002;
Barthe and Pelletier, 2007), butanol (Liste and Alexander, 2002; Swindell and Reid, 2006),
and highly pressurized gas (supercritical fluid extraction, (Librando et al. , 2004)) have been
used for this purpose. Other approaches focussed on the freely dissolved concentrations in
the pore water. Freely dissolved aqueous concentrations can be determined by equilibrium
dialysis (McCarthy and Jimenez, 1985), gas purging (Resendes et al. , 1992), non-
77
equilibrium passive samplers such as semipermeable membrane devices (SPMD) (Sproule
et al. , 1991; Booij et al., 1998), and equilibrium passive samplers such as
poly(oxymethylene) (POM) (lonker and Koelmans, 2001), polymer-coated glass sheets or
fibers (Mayer et al. , 2000; Heringa and Hermens, 2003), and low-density polyethylene
(LDPE) (Booij et al. , 2003; Adams et al. , 2007). Equilibriurn passive sarnplers are left in
contact with contaminated soil or sediment slurry, and freely dissolved aqueous
concentrations can be calculated with compound-specifie passive sampler-water partition
coefficients.
In a previous study on chemical availability of P AHs in low and highly contarninated
sediments, we compared solid/liquid bulk extraction methods with bioaccumulation in
worms (Barthe and Pelletier, 2007). Results indicated strong differences in distributions of
P AHs obtained using different extraction techniques. Butanol (BuOH) was found to extract
mu ch more PAHs than worms (Nereis virens and Lumbriculus variegatus) can
bioaccumulate in their tissues at steady-state, and hydroxypropyl-~-cyclodextrin (HP CD)
underestimated the bioavailability of P AHs, especially for low molecular weight P AHs. On
the other hand, the high molecular weight surfactant Brij® 700 (B700) proved to be a good
predictor for P AHs in aged highly contaminated sediments (Barthe and Pelletier, 2007).
In the present study we extended our investigation by studying the efficiency of
passive solid samplers polyoxymethylene (POM) strips (lonker and Koelmans, 2001) and
polydimethylsiloxane (PDMS) silicon tubing using sediment samples already described in
our previous work (Barthe and Pelletier, 2007). Results are compared with the previously
measured B700 liquid/solid extraction and worm bioaccumulation. Moreover,
78
bioaccumulation was estimated us mg the biota-sediment accumulation factor (BSAF)
(Boese et al. , 1996) as an instantaneous measurement of normalized tissue/sediment P AH
concentrations.
MATERIALS & METHODS
Chemicals
Hexanes (liquid chromatography quality) and acetone (high resolution gas
chromatography) were purchased from VWR Ltd (Mississaga, Canada). n-Heptane was
purchased from Merck (Darmstadt, Germany). InternaI PAH standard dio-phenanthrene
(d 10-PHE) was obtained from Cambridge Laboratories (Sweden). Additive-free polymer
materials were used, including medical-grade PDMS silicon tubing (core diameter 750 Ilm,
thickness 200 Ilm) from A-M systems, Inc (Carlsborg W A, USA) and polyoxymethylene
(POM) of 55 Ilm thickness from Astrup AS, Oslo, Norway (POM-55 , obtained in ~ l kg
cylinder-shaped blocks and sliced on a lathe equipped with a high-precision razor blade).
POM-55 consisted of C-POM, which is a copolymer of (CH20)n produced from trioxane
and other monomers.
Sediments
Marine sediments were collected during years 2003 and 2004 in Kitimat Fjord (BC,
Canada), in the St-Lawrence River and Estuary and in the Saguenay Fjord (QC, Canada).
79
Freshwater sediments came from the St-Louis River near Montreal (Qc, Canada). Samples
were stored in a freezer at -20 oC until analysis.
80
Table 1. Identification of samples for low and highly contaminated sediments. Total PAHs
in sediment (Csed) inc1ude: fluorene (FLU), phenanthrene (PHE), fluoranthene (FLT),
pyrene (PYR), benzo[a]anthracene (BaA), chrysene (CHR), benzo[b]fluoranthene (BbF),
benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), dibenzo[a,h]anthracene (DBA), and
benzo[g,h,i]perylene (BPE). TOC = Total organic carbon
Samples analyzed Csed (Ilg g-I w.w.) TOC (%)
Minette Bay (Kitimat, BC)a 0.06 ± 0.02 0.04
-a SLR upstream (St-Louis River)b 0.3 ± 0.1 5.1 v ..... ro .5 CIl ..... E s::::: Hospital Beach (Kitimat, BC)a 0.3 ± 0.1 0.08 v ro E ..... s:::::
~ 0 () v
EGSL04-07 (St. Lawrence Estuary)a 0.4 ± 0.1 1.1 ~ CIl
0 ~
EGSL04-13 (St. Lawrence Estuary)a 1.1 ± 0.2 0.7
SLR downstream (St-Louis River)b 25 .5 ± 2.6 2.5
-a SLR outflow (St-Louis River)b v 279 ± 29 2.0 ..... ro
s:::::
E CIl ..... s::::: Scow Grid (Kitimat, BC)a ro Q) 811 ± 155 2.4 .....
s::::: .5 0 () -a >-. Q)
Scrubber (Kitimat, BC)a CIl 4442 ± 538 41.0 ::a O!l
tE B Lagoon (Kitimat, BCt 5723 ± 190 29.9
a Marine sediments tested with Nereis virens;
b Freshwater sediments tested with Lumbriculus variegatus .
81
Bioaccumulation studies
Bioaccumulation tests used Nereis virens for manne sediments and Lumbriculus
variegatus for freshwater sediments as previously described (Barthe and Pelletier, 2007).
Tests were conducted in triplicate for 28 d, which is a sufficient time to ensure a steady-
state tissue concentration in both species, at temperature ranging between 4 and 6 oc. One
adult polychaete N. virens (3-5 g) was placed in each of the triplicate 1 L glass beaker with
300 mL of sediment to be tested and 700 mL of flow-through seawater with aeration. Sixt Y
(0.4 to 0.5 g) adult oligochaetes L. variegatus (approximately 2 cm in length) were placed
in each of the triplicate 500 mL glass beaker with 200 mL of freshwater sediment to be
tested and 300 mL of flow-through dechlorinated tap water with aeration. Worms
(polychaetes and oligochaetes) were sampled at seven times (0, 1, 3, 5, 7, 14, and 28 day)
during the exposure period of 28 days, (Barthe and Pelletier, 2007). A careful examination
of the results indicated no differences in uptake and elimination behavior for both worm
species and bioaccumulation results were discussed together.
Analysis of worm tissues
Total lipids were determined gravimetrically following the method described by
Folch et al. (1957). Nereis virens and Lumbriculus variegatus showed an average lipid
content of 8.33% and 10.45%, respectively. PAH analysis in worms followed the technique
previously described (Barthe and Pelletier, 2007). Briefly, samples (200 mg d.w.) were
extracted with 5 ml ofhexane:acetone (50:50 v/v) using an ultrasonic bath for 30 min. Each
sample was then shaken for 3 h and returned to the ultrasonic bath for 30 min. The
82
extraction volume was reduced to 0.5 mL, solvent exchanged to acetonitrile and reduced to
0.2 mL. The extracts were analyzed by liquid chromatography (LC) with fluorescence
detection.
Sediment characterization
Total PAHs, chemically available PAHs and total organic carbon (TOC) content were
determined in triplicate for the sediment samples using the same procedure as previously
reported (Barthe and Pelletier, 2007). Briefly, samples (1.0 g d.w.) were extracted with 10
mL of DCM. Each sample was then shaken for 16 h. The extraction volume was reduced to
0.5 mL, solvent exchanged to acetonitrile and volume reduced to 1 mL. The chemically
available PAHs (CB700) were extracted by B700 solution. Briefly, samples (1.0 g w.w.)
were extracted with 10 mL of B700 solution. Each sample was then shaken for 16 h and
then solvent exchanged to DCM. The extraction volume was reduced to 0.5 mL, solvent
exchanged to acetonitrile and volume reduced to 1 mL. The extracts were analyzed on a
Shimadzu LC-IOAD® pump fitted with a SupelcosiJTM LC-PAH column (25 cm x 3 mm)
and Spectra SYSTEM FL3000 fluorescence detector (HPLC-Fl). These two methods have
been shown to give good recovery (85-97%) ofPAHs. TOC contents were determined by a
carbon analyzer (Costech, Elemental Combustion System, CHNS-O, ECS 4010) after
acidification of the dry and ground sample with HCI (10 %) and a digestion at 80 oC for 15
h to eliminate carbonates (Barthe and Pelletier, 2007). Ali the chemical evaluation was
performed prior to the exposure of the organisms to the sediment.
Sediment characteristics for low and highly contaminated samples are given in Table
1. Low contaminated samples show PAH concentrations below 1 Jlg g-l (wet weight)
83
whereas highly contaminated samples contained PAHs ranging between 15.9 and 3605 /lg
gol (wet weight). TOC is usuaIly higher in highly contaminated samples except for St-Louis
River upstream (SLR upstream) sample located in a pristine area without industrial PAH
sources.
Freely dissolved concentrations/sediment sorption experiments
AU sorption studies were carried out in triplicate in 50 mL aIl glass flasks . Dry
sediment (3-10 g), passive sampler (100-300 mg), 100 mg NaN3 (biocide), NaCI for the
marine sediments (1.25 g, i.e., 2.5%, providing a constant ionic strength comparable to the
in situ conditions at the sampling sites), and Millipore water (alpha-Q, 50 mL) were shaken
end-over-end (6 rpm; 21 d). It was recently shown that 10 d equilibration time suffices for
PAHs in a sediment-water system with POM or silicon (G. Cornelissen, personnal
communication). After equilibrium time, the passive samplers were removed from the
system and cleaned with wet tissues.
P AH extraction from passive samplers
The clean passive sampler strips and tubing were extracted by horizontal shaking
(150 rpm; 72 h) with 15 mL of heptane in the presence of an internaI standard (400 ng and
40 ng dlO-PHE for highly and low contaminated sediments, respectively) (Cornelissen and
Gustafsson, 2004). The samples were eluted through a silica micro-column with 15 mL of
heptane. The extracts were analyzed by gas chromatography and mass spectrometry
(GC/MS) equipped with an 30 m x 0.25 mm DB-5 fused silica column (film thickness 0.25
/lm), and a ThermoFinnigan Polaris Q mass spectrometer in electron impact mode (Er, 70
84
eV). Quality control work showed recovery data ranging from 89.5 ± 5.6% (mean ±
standard deviation) to 98.7 ± 2.5%. Variability of analytical results was estimated to ± 15%
(n = 5) and the limit of detection was estimated at 0.001 ng /lL- l.
RESUL TS AND DISCUSSION
BSAF calculation
BSAFs (Biota-Sediment Accumulation Factors) were calculated in worms for each
analyzed PAH from the experimental data using Equation (1):
BSAF = C/iPid worms C
TOC
(1)
where Clip id is the lipid-normalized concentration of PAHs in the organisms (/lg g-l W.W.
lipid) and Croc is the TOC-normalized concentration in the sediment (/lg g-l W.W. organic
carbon). Median BSAFworms values ranged from 0.0008 to 0.22 in nine out often sediments
(Table 2), with the exception of SLR Upstream where the median reached 2.53. This higher
BSAFworms value might be attributed to the use of L. variegatus for this freshwater sediment
sample, but the tendency of higher bioaccumulation with L. variegatus was not confirmed
with SLR Downstream and SLR Outflow samples. Values of BSAFs in highly
contaminated samples were often two to four orders of magnitude lower than the theoretical
value of 1 (DiToro et al. , 1991). In low contaminated samples, ratios between measured and
theoretical BSAFs (Table 2) were as high as 15, whereas in highly contaminated samples,
which have been collected in the vicinity of aluminum smelters, the ratios reached 1466.
85
In order to include the effect of strong sorption to CGC (carbonaceous geosorbent
carbon) on bioaccumulation, BSAFs can also be estimated on the basis of freely dissolved
aqueous concentrations C wJree, determined using the POM and silicon tubing passive
sampler methods (Comelissen et al. , 2006) (Eq. 2).
C C exlracl 1 - x-w,free - K
ms s (2)
where Cextract is the amount in the extract (ng), ms is the mass of the sampler (kg) and Ks is
the sampler-water partition coefficient (log K; L kg-l) as determined.
In this approach, the overall TOC sorption, not only amorphous orgamc carbon
(AOC) but also CGC, is taken into account. BSAFest,free values (reported as BSAFest,POM and
BSAFest,silicon) were estimated on the basis of measured chemical parameters according to
the Equation (3):
KI· ·dC fi BSAF. = 'P' w, ree free C
TOC
(3)
where K lipid is the lipid-water partition coefficient (mL g-l) and can be approximated as
being equal to the octanol-water partition coefficient [KowJ (Mackay et al. , 1991) for the
specific PAHs and CwJree is the freely dissolved aqueous concentration (Ilg mL-1)
determined by the POM or the silicon tubing method. Chemically estimated BSAFest,free
values were compared with BSAFworms values by calculating BSAFest,free/BSAFworms ratios
(reported as BSAFest,POM/BSAFworms and BSAFest,silicon/BSAFworms) (Table 2). Most low
contaminated sediments provided relatively low ratios (aH below 0.7) for POM whereas
highly contaminated sediments produced ratios around 1 in most cases. Consistently,
86
silicon delivered lower ratios with low contaminated sediments and higher ones with
sediments loaded with PAHs. However, POM (0.3-1.4) and silicon (1.4-20) predicted
values are much closer to measured values than theoretical ones (0.4-1500). According to
Table 2, Cw estimates by the POM are a factor of 4-20 lower than those measured by silicon
tubing (data not shown). The differences between Cw estimated by the POM and the silicon
could be explained by the fact that the silicon tubing-water coefficient partition (Ksil) for aIl
the studied PAHs were greater than the POM-water coefficient partition (KpOM). PAHs
could be more attracted by PDMS than POM. The difference between the sampler material
is weIl illustrated by Rusina et al. (2007), who demonstrated that the diffusion coefficient
(D) of silicon tubing was greater than POM one (log Dsil ranged between - 9.95 and - 11.35
whereas log DpOM were below -16).
Using previously reported data (Barthe and Pelletier, 2007) for surfactant B700,
BSAFest,B700 were estimated according to Equation (4):
BSAF. - K /iPidC
B700 8 700 - C
TOC
(4)
where CS700 is the concentration of PAHs (Ilg mL- 1) deterrnined by the B700 method. The
BSAFest,B700 values were compared to BSAFworms by calculating BSAFest,B700/BSAFworms
ratios (Table 2). Again, ratios with low contaminated sediments were consistently lower
than those obtained with highly contaminated sediments with an average value of 1.17
(Table 2) .
Table 2. Biota-sediment accumulation factors (BSAF) deterrnined for ail PAHs in aU sediment samples as weil as ratios
between theoretical and empmc BSAFs (BSAFtheoreticaI/BSAFwonns) and between BSAFs calculated on the basis of
concentrations extracted by the different methods and empirical ones (BSAFest,POMIBSAFwonns, BSAFcst,s iliconlBSAFwonns and
BSAFest,B70oIBSAFwonns)
BSAFwonns BSAFtheoreticalal BSAFpOMI BSAFsiliconl BSAFB700i
BSAFwonns BSAFwonns BSAFwonns BSAFwonns
Minette Bay 0.16 (0.07-0.19)6 6.19 (5.23-30.9) 0.33 (0 .15-0.62) 3.38 (l.49-34.6) 0.18 (0.l6-0.28)
SLR Upstream 2.53 (1.43-4.43) 0.39 (0.24-0.69) 0.27 (0 .17-3 .85) 1.40 (0.79-22.9) 0.09 (0.06-0.22)
Hospital Beach 0.02 (0.02-0.06) 41.9 (17.9-54.7) 0.46 (0.34-0.71) 6.63 (4.72-21.4) 0.47 (0.19-0.60)
EGSL04-07 0.22 (0.11-0.37) 5.62 (2 .75-12.7) 0.33 (0.24-2.08) 2.4 7 (1.43-6.28) 0.14 (0.09-0.34)
EGSL04-13 0.06 (0.03-0.19) 15.0 (5 .36-35.9) 0.68 (0.49-8.38) 2.36 (l.88-25 .3) 0.49 (0.36-0.52)
SLR Downstream 0.02 (0.01 -0.03) 40.9 (30.0-348.2) 0.95 (0.46-2.43) 9.73 (5 .02-23.4) l.18 (0.40-1.47)
SLR Outflow <0.01 111 (76-347) 1.01 (0.61-1.16) 20.2 (7.5-73 .7) 0.99 (0 .39-1.12)
Scow Grid <0.01 1466 (1053-7005) 0.95 (0.28-2.13) 4.46 (2.02-6.05) 0.94 (0 .39-1.15)
Scrubber 0.l2 (0.08-0.15) 8.34 (6.59-12.04) 1.04 (0.29-2.09) 5.95 (1.79-9.92) l.26 (0.49-1.32)
B Lagoon 0.01 (0.01-0.02) 100 (44-207) l.37 (0.93-2.03) 7.72 (3.74-9.32) 1.50 (0.62-l.74)
a BSAFtheoretical value = 1 (DiToro et al. , 1991); 6 Median (interquartile ranges)
00 -.J
88
The low freely dissolved PAH concentrations deterrnined by POM and silicon and the
low concentrations extracted by the B700 method in these sediments could be explained by
strong sorption to GCG (Comelissen et al. , 2006), which imply the low BSAFs observed
for these sediments.
From a theoretical point of view, BSAF should approach unity for conserved organic
contaminants. Bioaccumulation of hydrophobic contaminants with BSAF of approximately
one has been frequently observed. For example, Ankley et al. (1992) demonstrated BSAFs
of approximately one for both laboratory and field populations of oligochaetes exposed to a
wide variety of polychlorinated biphenyl congeners in Fox River/Green Bay (WI, USA)
sediments. Biota-sediment accumulation factors of less than unit y, which appears to be the
case for most sediments, and particularly those with a high P AH levels, may therefore
represent direct indication of limited bioavailability (or altematively be a function of
elimination or breakdown of contaminants after uptake). So, it IS possible that
biodegradation of P AHs has influenced our results to a certain extent (Leppanen and
Kukkonen, 2000; J0rgensen et al., 2005). Lu et al. (2003) demonstrated that differences
between BSAF of desorption-resistant phenanthrene (0.59) and BSAF of reversibly sorbed
(labile) phenanthrene (1.2) were not due to differences in contaminant elimination and
metabolism nor in desorption kinetics nor in health of the worms but were due to a
reduction in the bioavailability of the contaminants in the sediment. Recently, Kreitinger et
al. (2007) used supercritical carbon dioxide extraction as a predictor of polycyclic aromatic
hydrocarbon bioaccumulation and toxicity by earthworms in manufactured-gas plant site
soils and also concluded that low BSAFs were related to low availability of contaminants.
89
In our case, we believe the low values of BSAFworms in highly contaminated samples can
mainly be attributed to strong sorption to aluminum smelter residues weIl characterized by
Breedveld et al (2007).
Earlier observations by Kraaij et al. (2003) suggested that freely dissolved pore-water
concentrations (CwJree) could be obtained from direct measurements using solid phase
microextraction (SPME) even at low concentrations of HOCs in the pg L-1 range.
Cornelissen et al. (2006) found that CwJree measured by the POM method was a good
predictor for BSAFs of native PAHs in three contaminated sediments (9 to 161 mg kg- I)
and for two organisms (Nereis diversicolor and Hinia reticulata) which is in agreement
with our results. These results are in agreement with our previous paper (Barthe and
Pelletier, 2007) where it was observed that low and highly contaminated sediments
presented differences in the extraction of the bioavailable P AHs by different che mi cal
techniques. Passive samplers also illustrated a difference between low and highly
contaminated sediments.
Correlation between BSAFs
In a second step, we examined the relationship between mean BSAFworms (for aIl
determined P AHs) at steady-state and the mean BSAF est,POM, BSAF est,s ili con or BSAF est,B700
as predicted by each method for both groups of samples pooled together (Fig. l , left
panels). The mean BSAFworms are reasonably weIl correlated with mean BSAFest,B700 (R2 =
0.93), BSAFest,POM (R2 = 0.71) and BSAFest,s ilicon (R2
= 0.88) wh en sample SLR upstream is
excluded for aIl three samplers and when the Scrubber sample is excluded for POM and
Silicon passive samplers. As already observed in Table 2, the BSAFworms of SLR upstream
90
is out of range compared with other samples because this pristine freshwater sample
combines a high percentage of TOC (5.1 %) (Table 1) with a very low PAH content (0.2 Ilg
g-l) which lowers the value of Croc in equation 1 and increases BSAFworms' The scrubber
sample is a very particular case because this sample is mainly composed of aluminum
smelter residues without natural sediment as described by Breedveld et al (2007). Only
surfactant extraction B700 provided a consistent result for that highly contaminated sample.
The second relationship to be examined was between individual BSAFworms (for each
analyzed P AH and both worm species) at steady-state and their predicted BSAF est,POM,
BSAFest,silicon or BSAFest,B700 only for low contaminated samples (Fig. 1, right panels).
These graphies illustrate the po or relationship existing between BSAF determined by
chemical methods and by worms for ail analyzed P AHs. Although mean BSAFs can be
consistently predicted by ail three samplers (left panels), individual BSAFs are not
adequately predicted, particularly for low contaminated samples.
91
Fig. 1. Mean of BSAFest,8700 (A), BSAFest,roM (B), and BSAFest,sili con (C) as a function of
mean BSAFworms over 28 days for low and highly contaminated sediments (left panels).
BSAF est,8700 (D), BSAF est,rOM (E), and BSAF est,silicon (F) of individual P AH as a function of
BSAFworms over 28 days for low contaminated sediments (right panels). The dotted lines
represent a hypothetical 1: 1 correlation. Scrubber sample is identified with letter s.
0.5
0.4
g ':!t'à; 0.3 en (]J
~ 0.2 <Il
::2:
0.1
A
s
•
0.0 41'----,---,---,---,___-___,
0.5
0.4
3 LLO. 0.3 ëJj (]J
c m 0.2 ::2:
0.1
0.0 0.1
B
0.2 0.3 ~an BSAFworms
• R2 = 0.7107
/
/
0.4 0.5
o~-_,--_,--._--,___-___,
0.0 0.1
2 c
0.2 0.3 ~an BSA F worms
R2 = 0.8798 •
0.4
s •
0.5
o·~~-,---,_-_,--_,--~
0.0 0.1 0.2 0.3 0.4 0.5 ~an BSA F worms
R CD
LL o:! en (]J
3 0.
LL ëJj (]J
2
2
•
•
2
D
E
• •
F •
• •
•
BSAF worms
•
BSAFworms
/ /
•
•• 2
•
2
•
o~~-----__,_------__,
o 2
92
To gain a better understanding on the capacity of B700 and POM methods to predict
individual PAR bioaccumulation In worms, the ratio of individual PARs
BSAFest,B7oo/BSAFworms and BSAFest,POM/BSAFworms were plotted against PAR octanol-
water partitioning coefficients (log Kow) for low contaminated (Fig. 2 top panels) and
highly contaminated sediments (Fig. 2, bottom panels). The ratio of individual PARs
BSAFest,s iliconlBSAFworms plotted against log Kow was not presented here because it has been
previously shown that silicon tubing was not a good predictor of the PAR bioavailability. A
ratio approaching 1 indicates the ability to predict PAR bioavailability based on B700 or
POM extractions, whereas a ratio <1 indicates an underestimation by B700 or POM
extraction of the biota-to-sediment accumulation factor and a ratio > 1 indicates an over-
prediction of the biota-to-sediment accumulation factor by the B700 or POM method. For
surfactant B700, low contaminated sediments show low ratios for low log Kow «5.8) but
high ratios (3.9 to 45) for high log Kow (>5.8). Righly contaminated sediments presented a
relatively similar picture as PARs with log Kow <5 exhibit a ratio <1, those with log Kow
between 5 and 5.8 exhibit a ratio almost equal to 1, and high ratios (6 to 29) are observed
for PARs with log Kow >5.8 (Fig. 2).
For POM sampler, low contaminated sediments presented a pattern similar to B700
except for fluorene (FLU) which has a ratio >5 (Fig. 2). Rowever, highly contaminated
sediments show a quite different pattern from B700 with three PARs with ratios close to
one (FLU, pyrene (PYR), and chrysene (CRR)), four PARs with ratios slightly above one
(phenanthrene (PRE), fluoranthene (FL T), bz[ a ]pyrene (BaP), and bz[b ]fluoranthene +
93
bz[k]fluoranthene (BbkF)), one PAH with ratios under one (bz[a]anthracene (BaA)) and
one PAH with a very high ratio (bz[ghi]perylene (BPE)) (Fig. 2).
Fig. 2 illustrates contrasts and similarities between low and highly contaminated
sediments for both B700 and POM methods. Almost aIl the low molecular-weight (LMW)
PAHs (log Kow <5.8) exhibit a ratio <1 in low contaminated samples indicating that worms
bioaccumulated more LMW P AHs in their tissues than normally expected from samplers.
As worms can digest marine organic matter, they can extract LMW P AHs already
bioaccumulated by living and dead organisms present in sediment which means a better
access to sm aIl P AH molecules particularly in low contaminated sediments were it is
assumed that most P AHs are directly associated with organic matter freshly derived from
biological activity. In contrast, both techniques are overestimating the availability of high
molecular-weight (HMW) PAHs (log Kow >5.8) for aIl sediments tested. It may indicate
that both techniques do not take in account biological factors limiting bioaccumulation of
HMW P AHs that seems to be related not only to log Kow but also to the molecular size of
the compounds. As an example, the molecular volume (A3) of fluorene is about 1.5 times
smaller than the one of dibenzoanthracene.
Fig. 2. Relationship between PAR octanol-water partitioning coefficient (log Kow) and the ability to predict PAH BSAF using
B700 and POM methodology. The studied PARs are: fluorene (FLU), phenanthrene (PRE), fluoranthene (FLT) , pyrene
(PYR), benzo[a]anthracene (BaA), chrysene (CRR), benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF),
benzo[a]pyrene (BaP), dibenzo[a,h]anthracene (DBA), and benzo[g,h,i]perylene (BPE).
Low contaminated sedim e n ts Low contaminated sed iments
1000 1000
100 J l 100
~ 10
l ! "6 ( 10 ~ f 1 :!:
....... ...... ... ..... .. ..
l î r î ~
~1 ..
0 . 1 ! ~! 0.1 ~ î
0 .0 1
0 .0 0 1 0 .01
4 5 6 7 8 4 5 6 7 8
Log K o w Log Kow
H ig hly contaminated sed im e nts Hig hly contaminated sedim e nts
10 00 1000
100 100 J l • ~ ! l ~ ; l l ( 10
~ 10 l 4 Y:!J:l . . . . . . . . . . . . . . . . . . . J
I1il ·· 1 !:ï '1 ~t 0. 1 ~ ~i
l 0 . 1
0 .0 1
0 .0 0 1 0.01 I..D 4 5 6 7 8 4 5 6 7 8 ~
Log Ko .. Log Ko ..
95
CONCLUSION
The two passlve samplers were found to act differently. The silicon sampler
overestimated the availability of P AHs in all studied sediments whereas the POM method
provided results quite similar to the solid/liquid extraction using high molecular weight
Brij®700. However, both methods poorly predicted availability in low contaminated
sediments where biological factors (digestible organic matter) become important.
Bioavailability of total PAHs was correctly predicted by POM and B700 in highly
contaminated aluminum smelter sediments. A closer examination of individual P AH results
indicated that both techniques overestimated the availability of large molecules with log
Kow >6 suggesting the presence in worms of a biological mechanism limiting uptake of
larger P AHs.
ACKNOWLEDGEMENTS
This study was supported by a grant from the Canadian Research Chair in Molecular
Ecotoxicology (E.P.) and from the Natural Science and Engineering Research Council of
Canada (NSERC) - Collaborative Research Development Pro gram (CRD). Samples have
been provided by Alcan Ltée. This paper is a contribution of the Quebec-Ocean Network.
The Research Council of Norway contributed to MB 's stay in Norway through a Strategie
Institute Pro gram (SIP).
96
REFERENCES
Adams, RG. , Lohmann, R. , Fernandez, L.A., MacFarlane, J.K. , Gshwend, P.M. 2007.
Polyethylene devices: passive samplers for measuring dissolved hydrophobic orgamc
compounds in aquatic environments. Environ. Sci. Technol. 41 , 1317-1323.
Ankley, G.T. , Cook, P.M., Carlson, A.R.; CalI, DJ., Swenson, lA., Corcoran, H.F. ,
Hoke, RA. , 1992. Bioaccumulation of PCBs from sediments by oligochaetes and fishes:
comparison of laboratory and field studies. Cano J. Fish. Aquat. Sci. 49, 2080-2085.
Barthe, M. , Pelletier, E. , 2007. Comparing bulk extraction methods for chemically
available polycyclic aromatic hydrocarbons with bioaccumulation in worms. Environ.
Chem. 4, 271-283.
Boese, B.L. , Lee, H. , Specht, D.T., Pelletier, l, Randall, R 1996. Evaluation ofPCB and
hexachlorobenzene biota-sediment accumulation factors based on ingested sediment in a
deposit-feeding clam. Environ. Toxicol. Chem. 15, 1584-1589.
Booij , K. , Shiu, W.Y. , Mackay, D. , 1998. Calibrating the uptake kinetics of
semipermeable membrane devices using exposure standards. Environ. Toxicol. Chem. 17,
1236-1345.
Booij , K. , Hofmans, H.E., Fischer, C.V. , Van Weerlee, E.M., 2003. Temperature-
dependent uptake rates of nonpolar organic compounds by semipermeable membrane
devices and low-density polyethylene membranes. Environ. Sci. Technol. 37, 361-366.
97
Breedveld, G.D. , Pelletier E., St. Louis R. , Cornelissen G., 2007. Sorption characteristics
of polycyclic aromatic hydrocarbons in aluminum smelter residues. Environ. Sci. Technol.
41 , 2542-2547.
Carroll, K.M., Harkness, M.R. , Bracco, A.A. Balcarcel, R.R. , 1994. Application of a
pemeantlpolymer diffusional model to the desorption of polychlorinated biphenyls from
Hudson River sediments. Environ. Sci. Technol. 28, 253-258.
Chang, M.-C. , Huang, C.-R. , Shu, H.-Y. , 2000. Effects of surfactants on extraction of
phenanthrene in spiked sand. Chemosphere 41 , 1295-1300.
Cornelissen, G. , van Noort, P.C.M. , Govers, H.AJ. , 1997. Desorption kinetics of
chlorobenzenes, polycyclic aromatic hydrocarbons, and polychlorinated biphenyls:
sediment extraction with Tenax® and effects of contact time and solute hydrophobicity.
Environ. Toxicol. Chem. 16, 1351-1357.
Cornelissen, G. , Gustafsson, O., 2004. Sorption of phenanthrene to environmental black
carbon in sediment with and without organic matter and native sorbates. Environ. Sci.
Technol. 38, 148-155.
Cornelissen, G. , Breedveld, G.D., Naes, K. , Oen, A.M.P., Ruus, A. , 2006.
Bioaccumulation of native polycyclic aromatic hydrocarbons from sediment by a
polychaetes and a gastropod: freely dissolved concentrations and activated carbon
amendment. Environ. Toxicol. Chem. 25,2349-2355.
Cuypers, C. , Pancras, T. , Grotenhuis, T. , Rulkens, W. , 2002 . The estimation of PAH
bioavailability in contaminated sediments using hydroxypropyl-~-cyclodextrin and Triton
X-IOO extraction techniques. Chemosphere 46, 1235-1245.
98
DiToro, D.M., Zabra, C.S., Hansen, DJ., Berry, W.J., Swartz, R.C. , Cowan, C.E. ,
Pavlou, S.P. , Allen, H.E., Thomas, N.A. , Paquin, P.R. , 1991. Technical basis for
establishing sediment quality criteria for nonionic organic chemicals using equilibrium
partitioning. Environ. Toxicol. Chem. 10, 1541-1583.
Doucette, W.J., 2003 . Quantitative structure-activity relationships for predicting soil-
sediment sorption coefficients for organic chemicals. Environ. Toxicol. Chem. 22, 1771-
1788.
Folch, J. , Lees, M. , Stanley, G. , 1957. A simple method for the isolation and purification
oftotallipids from animal tissues. J. Biol. Chem. 226, 497-509.
Heringa, M.B. , Hermens, J.L.M., 2003 . Measurement of free concentrations usmg
negligible depletion-solid-phase microextraction (nd-SPME). TrAC, Trends Anal. Chem.
22, 575-587.
Jonker, M.T.O. , Koelmans, A. A. , 2001. Polymethylene solid-phase extraction as a
partitioning method for hydrophobie organic chemicals in sediment and soot. Environ. Sei.
Technol. 35 , 3742-3749.
J0rgensen, A. , Giessing, A.M.B. , Rasmussen, LJ., Andersen, O. , 2005 .
Biotransformation of the polycyclic aromatic hydrocarbon pyrene in the marine polychaetes
Nereis virens. Environ. Toxicol. Chem. 24, 2796-2805.
Kelsey, J.W. , Kottler, B.D., Alexander, M. , 1997. Selective chemical extractants to
predict bioavailability of soil-aged organic chemical. Environ. Sei. Technol. 31 , 214-217.
99
Kennish, M.J. , 1997. Polycyclic aromatic hydrocarbons. In: Kennish, M.J. (Eds,).
Practical Handbook of Estuarine and Marine Pollution. CRC Press, Boca Raton, FL, USA,
pp. 141-175.
Kraaij , H. , Mayer, P. , Busser, F. , Van het Boischer, M. , Seinen, W. , ToUs, J. , Belfroid,
A.C. , 2003. Measured pore-water concentrations make equilibrium partitioning work - a
data analysis. Environ. Sei. Technol. 37, 268-274.
Kreitinger, lP. , Quifiones-Rivera, A., Neuhauser, E.F., Alexander, M. , Hawthorne, S.B.,
2007. Supercritical carbon dioxide extraction as a predictor of polycyclic aromatic
hydrocarbon bioaccumulation and toxicity by earthworms in manufactured-gas plant site
soils. Environ. Toxicol. Chem. 26, 1809-1817.
Lee II, H. , Boese, B. , Pelletier, l , Winsor, M. , Specht, D. , Randall , R. , 1989. Guidance
Manual: Bedded Sediment Bioaccumulation Tests. EPA/600/x-89/302. U.S. Environrnental
Protection Agency. Duluth, MN and Washington, DC, USA.
Leppanen, M.T., Kukkonen, lV.K. , 2000. Fate of sediment-associated pyrene and
benzo[a]pyrene in the freshwater oligochaetes Lumbriculus variegatus (Müller). Aquat.
Toxicol. 49, 199-212.
Librando, V. , Hutzinger, O., Tringali , G., Aresta, M. , 2004. Supercritical fluid extraction
of polycyclic aromatic hydrocarbons from marine sediments and soil samples.
Chemosphere 54, 1189-1197.
Liste, H.H. , Alexander, M., 2002. Butanol extraction to predict bioavailability of PAHs
in soil. Chemosphere 46, 1011-1017.
100
Lu, X., Reible, D.D., Fleeger, lW., Chai, Y. , 2003. Bioavailability of desorption-
resistant phenanthrene to the oligochaete llyodrilus templetoni. Environ. Toxicol. Chem.
22, 153-160.
Luthy, R.G. , Aiken, G.R. , Brusseau, M.L. , Cunningham, S.D., Gschwend, P.M. ,
Pignatello, J.1. , Reinhard, M. , Traina, S.1. , Weber, W.1. , Westall, lC. , 1997. Sequestration
of hydrophobic organic contaminants by geosorbents. Environ. Sci. Technol. 31 , 3341-
3347.
Mackay, D. , Shiu, W.Y., Ma, K.C. , 1991. Illustrated Handbook of Physical-chemical
Properties and Environmental Fate of Organic Chemicals, Vol. 1. Lewis Publishers, Boca
Raton, FL, USA.
Mayer, P. , Vaes, W.H.1. , Wijnker, F. , Legierse, K.C.H.M., Kraaij , H. , ToUs, l , Hermens,
J.L.M., 2000. Sensing dissolved sediment porewater concentrations of persistent and
bioaccumulative pollutants using disposable solid-phase microextraction fibers. Environ.
Sci. Technol. 34, 5177-5183.
McCarthy, lF. , Jimenez, B.D., 1985. Interactions between polycyclic aromatic
hydrocarbons and dissolved humic material: binding and dissociation. Environ. Sci.
Technol. 19, 1072-1076.
Resendes, l , Shiu, W.Y. , Mackay, D. , 1992. Sensing the fugacity of hydrophobic
organic chemicals in aqueous systems. Environ. Sci. Technol. 26, 2381-2387.
Rusina, T.P. , Smedes, F., Klanova, l , Booij , K. , Holoubek, L, 2007. Polymer selection
for passive sampling: a comparison of critical properties. Chemosphere 68, 1344-1351.
101
Sproule, J.W. , Shiu, W.Y. , Mackay, D. , Schroeder, W.H. , Russell , R.W., Gobas, F.A.P. ,
1991. Direct in situ sensing of the fugacity of hydrophobie chemicals in natural waters.
Environ. Toxicol. Chem. 10, 9-20.
Swindell, A.L. , Reid, B.l. , 2006. Comparison of selected non-exhaustive extraction
techniques to assess PAH availability in dissimilar soils . Chemosphere 62, 1126-1134.
U.S. Environmental Protection Agency. 2000. Methods for Measuring the Toxicity and
Bioaccumulation of Sediment-associated Contaminants with Freshwater Invertebrates. 2nd
edition. EPA 600/R-99/064, Duluth, MN and Washington, DC, USA.
Yeom, L.-T. , Ghosh, M.M., Cox, C.D. , Ahn, K.-H., 1996. Dissolution of polycyclic
aromatic hydrocarbons from weathered contaminated soil. Water Sei. Technol. 34, 335-
342.
CHAPITRE III
BIOACCUMULATION AND BIOTRANSFORMATION KINETICS OF
POLYCYCLIC AROMATIC HYDROCARBONS (PAHs) IN
SEDIMENTS WITH VARIABLE LEVELS OF CONTAMINATION
M. Barthe and É. Pelletier
103
Résumé
Dans cette étude, Lumbriculus variegatus et Nereis virens ont été exposés à des
sédiments de composition et de concentrations variables en hydrocarbures aromatiques
polycycliques (HAP). La capacité de biotransformation de L. variegatus et N. virens a été
déterminée sur 28 jours en suivant la concentration des métabolites de phase l du
phénanthrène et du pyrène: le 9-hydroxyphénanthrène (9-0H-PHE) et 1-hydroxypyrène (1-
OH-PYR). Nous avons déterminé les coefficients de toxicocinétique, les facteurs de
bioaccumulation (BAF), et les facteurs d'accumulation biote-sédiment (BSAF) qui sont les
BAF normalisés par rapport au contenu lipidique de l'organisme et au contenu en carbone
organique du sédiment. Les paramètres cinétiques (la constante du taux d'ingestion (ks) et
la constante du taux d'élimination (ke)) des HAP ont été comparés avec l'hydrophobicité de
ces HAP, exprimée par le coefficient de partition octanol-eau (log Kow). Les résultats
montrent que le log ks diminue avec l' augmentation de log Kow pour tous les sédiments
étudiés. La relation négative entre log ks et log Kow suggère que le taux de désorption du
composé à partir du sédiment était un facteur important dans l'accumulation par les vers.
La détermination de la proportion relative des métabolites par rapport au phénanthrène total
et au pyrène total dans les tissus de vers après une exposition de 28 jours indiquait la
présence de 24% de 9-hydroxyphénanthrène et 17% de 1-hydroxypyrène chez N. virens, et
de 18% de 9-hydroxyphénanthrène et 20% de 1-hydroxypyrène chez L. variegatus. D'après
ces importantes proportions en métabolites, nous présumons que les métabolites présents
dans les vers sont principalement dus à la métabolisation des HAP parents plutôt qu'à la
104
bioaccumulation à partir des sédiments où les métabolites étaient présents en faible
concentration.
Mots clés: HAP, sédiments marin et lacustres, bioaccumulation, métabolites, cinétiques de
biotransformation, hydroxy-phénanthrène, hydroxy-pyrène.
105
Abstract
In the present study, Lumbriculus variegatus and Nereis virens were exposed to field-
collected sediments of varying composition and concentration of polycyclic aromatic
hydrocarbons (P AHs). The biotransformation capability of L. variegatus and N. virens was
assessed over 28 days by following the concentration of 9-hydroxyphenanthrene (9-0H-
PHE) and 1-hydroxypyrene (l-OH-PYR), the phase l metabolites of phenanthrene and
pyrene, respectively. Toxicokinetic coefficients, bioaccumulation factors (BAF), and biota-
sediment accumulation factors ([BSAF] , BAF normalized to the organism lipid content and
sediment organic carbon content) were determined. The kinetic parameters (the uptake
clearance rate constant (ks) and the elimination rate constant (ke)) of PAHs were compared
to molecular hydrophobicity of these PAHs, expressed by the octanol-water partition
coefficient (log Kow). The results showed that log ks decreased with increasing log Kow for
aU studied sediment samples. The negative relationship between log ks and log Kow
suggests that the desorption rate of the compound from the sediment was an important
governing factor in the accumulation by the worms. The determination of the relative
proportion of metabolites toward total phenanthrene and pyrene in worm tissues after a 28-
day exposure indicated the presence of 24% of 9-hydroxyphenanthrene and 17% of 1-
hydroxypyrene in N. virens, and 18% of 9-hydroxyphenanthrene and 20% of 1-
hydroxypyrene in L. variegatus. According to these high proportions of metabolites, we
assumed that metabolites present in worms are mainly due to metabolization of the parent
P AH instead of bioaccumulation from sediment where metabolites were present in low
concentrations.
106
Keywords: P AH, manne and lacustrine sediments, bioaccumulation, metabolites,
biotransformation kinetics, hydroxy-phenanthrene, hydroxy-pyrene.
107
INTRODUCTION
Polycyclic aromatic hydrocarbons (PAHs) are common contaminants In manne
environments (Kennish, 1997) that accumulate and persist in sediments and can therefore
be taken up by sediment dwelling organisms (Landrum, 1989; Leppanen and Kukkonen,
1998; Loonen et al. , 1997; Weston, 1990). Thus, understanding the bioavailability of these
sediment-sorbed contaminants is important to determine their potential environmental risk.
In a general manner, the bioavailability of sediment-sorbed contaminants decreases with the
increase of the contact time (Landrum, 1989; McElroy and Means, 1988), where the contact
time corresponds to the elapsed time from the introduction of the contaminant in the
sediment. The contact time is also named aging time. The reduction of the bioavailable
fraction can be either due to the strong binding with the sedimentary organic carbon or to
the rapid metabolization of these compounds by the organisms (Pignatello, 1990).
Inversely, numerous studies using spiked sediments have indicated that sorne chemicals,
such as PAHs, did not show decrease of the bioavailability through time (Landrum, 1989;
Landrum et al., 1 992a). However, when comparing spiked and field polluted sediments, a
lower accumulation has been shown in the polluted sediments compared to the spiked ones
(Landrum et al. , 1992a; Varanasi et al. , 1985). These studies suggested that sorne
compounds need more contact time (more than one year) to reach equilibrium with
sediments. Due to the toxicity and the persistence of these compounds, it is important to
further study their bioaccumulation potential.
108
Bioaccumulation studies have been widely used to evaluate the bioavailability and the
ecological risk of bedded-sediments. These studies often incorporated a bioaccumulation
factor (BAF), which is defined as the ratio of the contaminant in the organism at steady-
state and the contaminant concentration in the environmental media (water, sediment,
etc ... ). The disadvantage of these studies is the important time needed for the organisms to
reach the steady-state for sorne compounds. Due to this drawback, toxicokinetic models, in
particular two-compartment models using first-order kinetics are commonly used to
de scribe accumulation and predict levels at steady-state under non-equilibrium conditions
(Landrum et al., 1992b; Lydy et al., 2000; Schuler and Lydy, 2001).
The present work was undertaken as part of a larger study exammmg the
sequestration mechanisms of P AHs in manne and lacustrine sediments with different
geochemical and environmental characteristics, and to study the bioaccumulation of P AHs
in order to correlate it with obtained sequestration data. The goal of the first part of this
study was to evaluate three different non-exhaustive solidlliquid extraction methods (n-
butanol , hydroxypropyl-p-cyclodextrin, and a surfactant solution of Brij700) for assessing
the availability of P AHs in contaminated sediments. This goal was achieved by comparing
results from 28-day uptake experiments by Nereis virens and Lumbriculus variegatus with
PAHs extracted by these three methods. Our results showed that an extraction procedure
involving the Brij700 was revealed to be useful particularly with sediments that show PAH
levels > 2 Jlg g-t and log Kow< 5.8 (Barthe and Pelletier 2007). The purpose of the second
part of the study was to test the efficiency of passive solid samplers, polyoxymethylene
(POM) strips and polydimethylsiloxane silicon tubing, to predict the bioavailability of
109
native PAHs in contaminated sediments. Results were compared to those of the first study.
The silicon sampler overestimated the availability of P AHs in aIl studied sediments
whereas the POM method provided results quite similar to the solid/liquid extraction using
high molecular weight Brij700. However, both methods poorly predicted availability in low
eontaminated sediments. A closer examination of individual P AH results indicated that the
POM technique overestimated the availability of large molecules with log Kow >6 (Barthe
et al., 2008).
The principal aim of the present study was to investigate the temporal patterns of
P AH accumulation for eleven sediments with variable levels of contamination in two
species of aquatic invertebrates (Lumbriculus variegatus and Nereis virens). Our specifie
objectives were to determine the toxicokinetic parameters describing the uptake,
elimination and BAF of the eight P AHs studied in sediments exhibiting large differences in
their chemical properties. Moreover, the ability of the organisms to metabolize
phenanthrene and pyrene was examined by the determination of selected hydroxy P AHs (9-
hydroxyphenanthrene and 1-hydroxypyrene).
MATERIALS & METHODS
Chemicals
Polycyclic aromatic hydrocarbons standards (method standards for waste water, 16
compounds 0.1 mg mL-') were purchased from Chromatographic Specialties Inc.
(BrockviIle, Canada). 9-hydroxyphenanthrene (9-0H-PHE) (teehnical grade) and 1-
110
hydroxypyrene (l-OH-PYR) (98% purity) were purchased from Sigma-Aldrich (Oakville,
Canada). Anthracene-d lO and fluoranthene-d lO were from Cambridge Isotope Laboratories
(Andover, MA, USA) at 98% purity. Acetonitrile (HPLC grade) and dichloromethane
(DCM) (Liquid Chromatography Analysis) were purchased from VWR (Mississauga,
Canada).
Sediments
Marine sediments were collected in the St-Lawrence River and Estuary and in the
Saguenay Fjord (QC, Canada) in July 2004 with a Van Veen grab and in the Kitimat Fjord
(BC, Canada) in September 2003 at low tide with shovel. The whole sediment was placed
into c1ean buckets and frozen at -20°C until analysis. Lacustrine sediments were collected
in the Saint-Louis River (SLR) (QC, Canada) in October 2003 with a hand corer. The
whole sediment was placed into c1ean buckets and frozen at -20°C until analysis. Total
P AHs, and total organic carbon (TOC) content were determined in triplicate for the
sediment samples using the same procedure as previously reported (Barthe and Pelletier,
2007). The sediment properties were reported in Barthe and Pelletier (2007). Low
contaminated samples (EGSL04-01; -07; 13; SLR Upstream; Minette Bay; Hospital Beach)
show PAH concentrations between 0.06 and 1.1 flg g-l (wet weight) whereas highly
contaminated samples (B Lagoon; Scrubber; Scow Grid; SLR Outflow; SLR DOwrIstream)
contained PAHs ranging between 25.5 and 5723 flg g-l (wet weight). TOC ranged between
2.0 and 41% and 0.04 and 5.1% inhighly and in low contaminated samples respectively.
111
Uptake experiments
Bioaccumulation tests used Nereis virens and Lumbriculus variegatus as previously
described (Barthe and Pelletier, 2007). Tests were conducted in triplicate for 28 d at
temperature ranging between 4 and 6°C.
Analysis of worm tissues and sediments
The PAR analysis in organisms and sediments followed the technique previously
described (Barthe and Pelletier, 2007). Metabolites in biological tissues and in sediments
were extracted with the same protocol used for the extraction of PARs in organisms.
Briefly, samples (200 mg d.w.) were extracted with 5 ml of hexane:acetone (50:50 v/v)
using an ultrasonic bath for 30 min. Each sample was then shaken for 3 h and returned to
the ultrasonic bath for 30 min. The extraction volume was reduced to 0.5 mL, solvent
exchanged to acetonitrile and reduced to 0.2 mL. Concentrations of phenanthrene, 9-
hydroxyphenanthrene, pyrene and 1-hydroxypyrene in sediments and in worms are
presented in Table l.
Table l. Concentration of phenanthrene (PHE), 9-hydroxyphenanthrene (9-0H-PHE), pyrene (PYR), and 1-hydroxypyrene
(l-OH-PYR) in sediments and in worms ().tg g-I w.w.) and percentage ofmetabolite compared to parent PAH.
PHEsed 9-0H-PHEsed PHEwomls 9-0H-PHEwonns %a PYRsed l-OH-PYRsed PYRwonns l-OH-PYRwonns %a
B Lagoon 213 11.8 0.55 0.15 22 660 0.02 4.1 l. 6 28
Scrubber 227 3.1 7.3 3.05 29 770 l.9 18.8 0.52 2.7
Scow Grid 48 0.56 0.13 0.05 28 95 0.003 0.46 0.19 29
SLR Outflow l.6 0.58 0.71 0.13 15 46 0.002 2.77 0.69 20
SLR Downstream 0.36 0.06 0.06 0.03 32 l.9 l.0 x 10-4 0.26 0.09 27
EGSL04-1 3 0.07 0.02 0.21 0.04 18 0.08 3.0 x 10-4 0.32 0.007 2.2
EGSL04-07 0.05 0.02 0.15 0.04 24 0.06 l.0 x 10-4 0.19 0.04 16
Hospital Beach 0.06 0.01 0.10 0.05 32 0.05 3.0 x 10-4 0.26 0.09 27
SLR Upstream 0.06 0.05 0.42 0.02 5.3 0.06 0.004 0.62 0.08 12
EGSL04-01 0.01 0.02 0.20 0.03 I l 0.02 l.0 x 10-4 0.31 0.03 9.6
Minette Bay 0.03 0.01 0.06 0.025 29 0.014 4.0 x 10-4 0.36 0.11 23
a % = (OH-PAHwonns*100)/(PAHwonns + OH-PAHwonns)
.......
....... N
113
Quantification
Metabolite analyses were carried out by liquid chromatography (LC) with
fluorescence detection. The apparatus consisted of a Rheodyne® in je ct or with a 20 ilL
injection loop, a Shimadzu LC-1 OAD® pump, a Supelcosil™ LC-PAH column (25 cm x 3
mm), and Spectra SYSTEM FL3000 fluorescence detector. AIl analyses were done at a
constant flow rate of 0.8 mL min- I with a mixture of nanopure water and acetonitrile as
mobile phase. The pump pro gram began at 75% acetonitri1e and increased to 95% in 10
min and then to 100% in the next 10 min, with a final plateau of 10 min. The cycle returned
to 75% acetonitrile after a total run time of 35 min. The excitation wavelength of the
fluorescence detector was settled at 244 nm during the first 3.4 min and then shifted to 344
mu until the end of the pro gram and the emission was detected at 370 nm during the first
3.4 min and then shifted to 394 nm until the end of the program. For quality control , a 0.5
Ilg mL- 1 9-0H-PHE and 1-0H-PYR mix was analyzed every 10 samples.
Fluorene (FLU), phenanthrene (PHE), fluoranthene (FL T), pyrene (PYR),
bz[a]anthracene + chrysene (BaA+CHR), benzo[bk]fluoranthene (BbkF), bz[a]pyrene
(BaP), bz[g.h.i]perylene (BPE), 9-0H-PHE and 1-0H-PYR were quantified.
Quality control work showed recovery data ranging from 89.5 ± 5.6% (mean ±
standard deviation) to 98.7 ± 2.5%. Variability of analytical results was estimated to ± 15%
(n = 5) and the limit of detection was estimated at 0.01 ng g-I (d.w. sediment).
114
Data treatment
Contaminant accumulation data were fitted to a first-order toxicokinetic, two-
compartment accumulation model to estimate both uptake and elimination for evaluation of
bioavailability (Kukkonen et al. , 2004; Spacie and Hamelink, 1983 ; Spacie et al. , 1995):
C = ksC, (1- - k,t ) worm s k e
e
(1 )
where Cworms is the concentration of the compound in the organism (Ilg g-I w.w.), Cs is the
concentration of the compound in the sediment (Ilg g-I d.w.) , ks is the uptake clearance rate
constant of the compound from sediment (g dry sediment g-I wet organism h-I), ke is the
elimination rate constant of the compound (h- I) in sediment, and t is time Ch). The model
assumes that the concentration in the sediment remains constant and there is no
biotransformation of the compound. The best-fit adjustable parameter values were found
using the sol ver function in Microsoft Excel® 7.0 (Table 2).
Table 2. Estimated wet weight-based uptake (ks) and elimination (ke) rate constants for the kinetics of polycyc1ic aromatic
hydrocarbons (P AHs) in polychaetes and oligochaetes.
Log B Lagoon Scrubber Scow SLR SLR EGSL04- EGSL04- Hospital SLR EGSL04-
K awa Grid Outflow Downstream 13 07 Beach Upstream 01
FLU 4.18 ks 3.0 x 10.4 0.0 1 1.6 x 10-4 5.0 x 10-3 2.0 x 10-3 0.21 0.20 0.46 0.13 0.97
kc 0.02 0.01 0.02 0.02 0.01 0.08 0.01 0.01 0.03 0.01
PHEN 4.52 ks 3.8 x 10-3 5.0 X 10-4 8.34e-5 3.0 x 10-3 2.0 X 10-3 0.33 0.08 0.04 0.07 0.78
ke 0.15 0.01 0.03 0.01 0.0 1 0.11 0.03 0.03 0.0 1 0.04
FLT 5.33 ks 1.0 x 10-4 5.0 X 10-4 8.41 e-5 5.0 X 10-4 1.0 X 10-3 0.07 0.07 0.88 0.56 0.53
ke 0.02 0.02 0.02 0.0 1 0.01 0.04 0.0 1 0.15 0.02 0.03
PYR 5. 18 ks 8.6 x 10-5 7.0x 10-4 4.0 X 10-4 5.0 X 10-4 2.0 X 10-3 0.2 1 0.05 0.50 0.14 0.82
kc 0.0 1 0.03 0.09 0.0 1 0.01 0.06 0.0 1 0.10 0.0 1 0.05
BaA+CHR 5.76b ks 2.1 X 10-5 2.0 X 10-4 2.69c-5 1.0 X 10-4 1.0 X 10-4 2.0 X 10-3 0.02 0.04 0.04 0.42
kc 0.01 0.01 0.05 0.01 0.01 0.0 1 0.02 0.02 0.01 0.02
BbkF 6.71 c ks 2.5 x 10-5 3.8 X 10-5 7.01 c-6 1.0 X 10-4 2.0 X 10-4 0.0 1 0.01 0.02 0.03 0.04
kc 0.01 0.02 0.0 1 0.0 1 0.01 0.0 1 0.01 0.0 1 0.0 1 0.02
BaP 6.31 ks 1.4 x 10-5 6.7 X 10-5 4.7x 10-6 1.0 x 10-4 8.4 X 10-5 3.0 X 10-3 0.01 1.0 x 10-3 0.54 0.01
kc 0.02 0.03 0.03 0.02 0.01 0.01 0.01 0.0 1 0.04 0.01
BPE 7.23 ks 2.3 x 10-5 4.0 X 10-5 4.5 X 10-6 1.0 X 10-4 8.5 X 10-5 6.0x 10-4 4.0 X 10-4 0.0 1 7.0 x 10-3 2.0 X 10-3
kc 0.02 0.01 0.0 1 0.01 0.0 1 0.01 0.02 0.0 1 0.01 0.0 1
a From Mackay et al. (1991); b mean of BaA and CHR Log Kow; C mean of BbF and BkF Log Kow
Minette
Bay
0.99
0.02
OAO
0.19
0.89
0.02
0.86
0.03
0.78
0.02
0.07
0.0 1
0.05
0.01
0.04
0.02
>-' >-' Vl
116
Bioaccumulation factors were determined either from the measured values at the end
of the exposures (BAFme) as the concentration in the organisms (Ilg g-I w. w.) divided by
the concentration in the sediment (Ilg g-I w. w.) or as calculated values from the
toxicokinetics (BAFca) as the ratio ofkslke to give the expected steady state BAF (Table 3).
The biota-sediment accumulation factor was calculated by normalizing BAFca by the
lipid content of the organisms and the organic carbon content of the sediments (Table 3).
The average lipid values for Lumbriculus variegatus and for Nere is virens were 10.45%
and 8.33%, respectively. The sediment organic carbon contents were reported previously
(Barthe and Pelletier, 2007).
Metabolite accumulation data were fitted to a first-order toxicokinetic, three-
compartment accumulation model to estimate uptake, metabolization and elimination rate
constants:
where CMO is the concentration of the metabolite compound in the organism (Ilmol g-I
w.w.), Cps is the concentration of the parent compound in the sediment (Ilmol il d.w.),
C pO is the concentration of the parent compound in the organism at to (Ilmol g-l ()
w.w.), C MOo is the concentration of the metabolite compound in the organism at to (Ilmol g-l
w.w.), k l is the uptake clearance rate constant of the parent compound from sediment (Ilmol
dry sediment il wet organism h- l), k2 is the metabolization rate constant of the parent
compound (h- l), k3 is the elimination rate constant of the metabolite compound (h- l) and t is
time (h). The best-fit adjustable parameter values were found using Lab Fit 7.2® (Table 4) .
Table 3. Bioaccumulation (BAF) and biota-sediment accumulation (BSAF) factors for the selected polycyclic aromatic
hydrocarbons.
B Scrubber Scow SLR SLR EGSL04- EGSL04- Hospital SLR EGSL04- Minette
Lagoon Grid Outflow Downstream 13 07 Beach Upstream 0 1 Bay
FLUa BAFmc 0.006 l.06 0.003 0.079 0.14 0.792 l.70 12.8 3 .03 4.87 33.6
BAFca 0.016 l.00 0.008 0.33 0 .14 2.62 14.3 38 .3 4 .81 88.2 45 .0
BSAF 0.06 4.92 0.002 0.06 0 .034 0.22 1.88 0.37 2.35 13.2 0 .22
PHEN BAFmc 0.003 0.03 0.003 0.45 0.17 2.80 3.00 l.70 6.90 19.9 2 .10
BAFca 0.03 0.03 0.003 0.43 0 .14 3.00 2.80 l.60 7.40 19.0 2. 10
BSAF 0.09 0.16 0.0007 0.08 0 .034 0.25 0.37 0.015 3.63 2.74 0.01
FLT BAFmc 0.007 0.03 0.004 0.045 0 .1 0 1.80 2.50 6.60 0.67 20.4 44.6
BAFca 0.006 0.03 0.005 0.05 0. 11 1.80 2.80 5.90 28 .0 20.4 52.3
BSAF 0.02 0.15 0.001 0.009 0 .025 0.15 0.64 0.056 13 .66 2.93 0.26
PYR BAFme 0.006 0.02 0.0048 0.061 0.14 4.10 3.40 5.20 1l.3 16.9 25 .1
BAFca 0.006 0.02 0.0046 0.062 0 .17 3.70 3.40 5.05 1l.7 16.7 26.1
BSAF 0.02 0.12 0.001 0.011 0 .039 0.31 0.45 0.048 5.69 2 .31 0. 12
BaA+CHR BAFme 0.002 0.02 0.0005 0.013 0.0 11 0.15 0.88 2.35 3.55 1.98 44.2
BAFca 0.002 0.02 0.0005 0.01 0 .011 0.25 1.35 2.43 3.90 20.0 45 .9
BSAF 0.008 0.08 0.0001 0.002 0 .002 0.02 0.18 0.023 1.91 2.88 0 .22
....... -..J
B Scrubber Scow SLR SLR EGSL04- EGSL04- Hospital SLR EGSL04- Minette
Lagoon Grid Outflow Downstream 13 07 Beach Upstream 01 Bay
BaP BAFme 0.001 0.002 0.0001 0.009 0.007 0.26 0.59 0.15 15.1 7.20 3.90
BAFca 0.001 0.002 0.0001 0.0066 0.007 0.25 0.61 0.11 13 .5 1.30 4.00
BSAF 0.003 0.012 5.0 x 10-5 0.001 0.002 0.02 0.08 0.001 6.58 0.19 0.02
BPE BAFme 0.001 0.005 0.0004 0.011 0.008 0.06 0.016 1.54 0.51 0.17 2.64
BAFca 0.001 0.005 0.0004 0.009 0.009 0.07 0.022 1.62 0.58 0.15 2.75
BSAF 0.004 0.024 0.0001 0.002 0.002 0.006 0.003 0.016 0.28 0.022 0.01
Median BAFme 0.002 0.02 0.002 0.03 0.06 1.01 1.29 2.08 3.29 6.05 15.6
BAFca 0.004 0.02 0.002 0.03 0.06 1.58 2.08 2.18 6.13 17.9 16.3
BSAF 0.015 0.10 0.0005 0.006 0.016 0.13 0.27 0.021 2.99 2.52 0.08
a FLU = fluorene, PHE = phenanthrene, FLT = fluoranthene, PYR = pyrene, BaA+CHR = bz[a]anthracene + chrysene , BbkF
= benzo[bk]fluoranthene, BaP = bz[a]pyrene, BPE = bz[g.h.i]perylene.
...... 00
Table 4. Estimated wet weight-based uptake (k l ) , metabolization (k2) and elimination (k3) rate constants for the kinetics of
hydroxy polycyclic aromatic hydrocarbons (OH-PAHs) in polychaetes and oligochaetes.
B Lagoon Scrubber Scow SLR SLR EGSL04- EGSL04- Hospital SLR EGSL04-
Grid Outflow Downstream 13 07 Beach Upstream 01
9-0H- k] 3.0 x 10-4 4.0 X 10-4 3.3 X 10-4 0.002 0.002 0.07 0.03 0.06 0.003 0.37
PHEN k2 0.35 0.07 0.02 0.02 0.02 0.03 0.08 0.02 0.007 0.02
k3 0.43 0.03 0.31 0.02 0.04 0.11 0.031 0.09 0.008 0.14
1-0H- k] 1.1 x 10-4 3.1 X 10-4 3.8 X 10-4 3.3 X 10-4 5.7 X 10-4 0 .004 0.01 0.28 0.003 0.02
PYR k2 0.04 0.04 0.08 0.03 0.01 0.002 0.02 0.14 0.02 0.03
k3 0.04 0.05 0.19 0.02 0.01 0.05 0.02 0.15 0.02 0.02
Minette
Bay
0.07
0.04
0.08
0.77
0.04
0.10
....... ....... \0
120
RESULTS
Bioconcentration of PAH in worms
A rapid uptake of P AHs by worms was indicated by the maximum tissue levels
measured within 7 days (168 h) of exposure for aU target compounds in aU sediment
samples and both worm speCles. Six sediments with different characteristics and
contamination levels were selected to iUustrate this point (Fig. 1). For most PAHs,
bioaccumulation showed a regular pattern although concentrations ranged from <0.05 ~g g-
1 in low contaminated sediments to as much as 20 ~g g-I in the highly contaminated
scrubber sample. Phenanthrene revealed peak levels within 72 h and 120 h in one highly
contaminated sediment (B Lagoon) (Fig. lA) and two low contaminated sediments
(EGSL04-07 and Hospital Beach), respectively (Fig. 1 D and F). Pyrene also revealed peak
levels within 72 h in one highly contaminated sediment (Scrubber) and in one low
contaminated sediment (Hospital Beach) (Fig. IF) . Fluoranthene also showed a particular
behavior in one low contaminated sediment (Hospital Beach) (Fig. 1 F).
121
Fig. 1. Uptake kinetics of fluorene ( . ), phenanthrene (0), fluoranthene (T ), pyrene ( ~ ) ,
bz[a]anthracene + chrysene (v), benzo[bk]fluoranthene (. ), benzo[a]pyrene (0), and
benzo[g.h.i]perylene (. ) in worms during the exposure period in B Lagoon (A), Scrubber
(B), SLR Downstream (C), EGSL04-07 (D), SLR Upstream (E), and Hospital Beach (F)
sediments. The curve corresponds to the nonlinear fit of the data using the model expressed
in Eq. 1. Note different scales for P AH concentrations.
A 1)
0.30
~ 5 ~ 0.25 ~ ~
~ el) 4 ~
el) 0.20 el) el)
2; 2; § .2 0. 15
g :. 2 l:: 0. 10
" " " " <) <)
" " 8 1 8 0.05
0.00 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time(h) Ti me(h)
B E
30 3.0
~ 25 ~ 2.5 ~ ~
~ el) 20 el) 2.0 el) el)
1/ 2; 2;
" 15 " 1.5 .12 0
g 10 g
1.0 " " " " <) <)
" " 0 5 0 0.5 ~ U U
0.0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time(h) Time (h)
C F
0.30 0.30
,..,. ,..,. ~ 0.25 ~ 0.25 ~ ~
~ el) 0.20 el) 0.20 el) el)
2; 2;
" 0. 15 " 0. 15 .12 .12 g 0. 10 g 0. 10 " " " " <) <)
" cS 0 0.05 0.05 U
0.00 0.00 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time(hl Time (hl
122
Uptake rate constants (ks) were calculated for individual PAHs using Eq. 1 (Table 2).
The relationship between uptake clearance rate constants and the molecular lipophilicity
expressed by log Kow is illustrated in Figure 2 (Top panel). The plots show that log ks
decreased with increasing log Kow for all studied sediment samples (R2 = 0.54 and 0.48 for
highly and low contaminated sediments, respectively).
Elimination rate constants (ke) were calculated for individual P AHs also using Eq.
(Table 2). The relationship between elimination rate constants and log Kow has been plotted
(Fig. 2 Bottom panel). The plots show that the elimination rate constant for the studied
worms also tends to decrease with increasing log Kow for both categories of sediment, but
with low correlation coefficients (R2 = 0.08 and 0.24 for highly and low contaminated
sediments, respectively).
The bioaccumulation factors (BAFs) were estimated from BAFca = ks /ke, and also
directly (BAFme) from P AH concentrations in worm tissues and sediment samples, (Table
3). A decrease of BAFca was observed with increasing log Kow for the low and highly
contaminated sediments (data not shown). Median BSAF values ranged from 0.0005 to
0.27 in ni ne out of eleven sediments (Table 3), with the exception of SLR Upstream and
EGSL04-01 where the medians reached 2.99 and 2.52, respectively. Values of BSAFs in
highly contaminated samples were often two to four orders of magnitude lower than the
theoretical value of 1 (DiToro et al., 1991).
123
Fig. 2. (Top) Plot of the log of uptake clearance rate constants, log ks, versus the log of the
hydrophobicity expressed by the octanol/water partition coefficient, log Kow, for highly (A)
and low (B) contaminated sediments. (Bottom) Plot of the log of elimination rate constants,
log ke, versus the log of the hydrophobicity expressed by the octanol/water partition
coefficient, log Kow, for highly (C) and low (D) contaminated sediments.
A B
log K"", log Kow
4 5 6 7 8 4 5 6 7 8 0 0
-1
-2 • -2
• t .:2 • R2 = 0.54 .:2 • • • Cl -3 Cl -3 • .2 .2 * -4 • • -4
• R2 = 0.48 -5 • -5
• • -6 -6
C D
log K"", log Kow
4 5 6 7 8 4 5 6 7 8 0 0
-1 • • •• • -1 • • • !i • • • : • • -2 • • S. • • 1 • -2 1 • • • • •
" " "" "" Cl -3 Cl -3 .2 .2
-4 -4
-5 R2 = 0.08 -5 R2 = 0.24
-6 -6
124
Biotransformation of P AHs in worms
Two hydroxy-PAHs, 9-hydroxyphenanthrene (9-0H-PHE) and I-hydroxypyrene (1-
OH-PYR), were used as proxies to quantify the presence of hydroxy-metabolites in worm
tissues and estimate metabolic activity of both worm species. After a 28-d exposure to
sediments, between 5.3 and 32%, with a median value of 24%, of total phenanthrene in
worm tissues was constituted by 9-0H-PHE and between 2.2 and 29%, with a median value
of 20%, of total pyrene in worm tissues was constituted by 1-0H-PYR (Table 1). Results
for three typical highly contaminated sediments and three low contaminated ones are
illustrated in Fig. 3 and 4 for phenanthrene and pyrene, respectively. The ratio 9-0H-
PHE/PHE was plotted alone with the PHE bioaccumulation curve for each sediment sample
(Fig. 3). Highly contaminated B Lagoon (Fig. 3A) provides a straightforward example
where ratio 9-0H-PHE/PHE increased quickly to a plateau as PHE was bioaccumulated in
the worm tissues within 24h. No significant differences (p > 0.05) were seen in the
foliowing days and weeks. In most other cases, the 9-0H-PHE/PHE ratios increased rapidly
in the first 24 to 72h, and then decreased and reached a plateau in the following days
(Fig.3B, C, and D). At steady state, the 9-0H-PHE/PHE ratio ranged for about 0.3 to 0.5
with the exception of SLR Upstream (Fig.3E) where the ratio never exceeded 0.1 even with
a quite high level of PHE in this low contamination sample. We found no evidence of a
relationship between 9-0H-PHE/PHE and PHE concentrations in aU samples studied.
125
Fig. 3. The ratio of 9-hydroxyphenanthrene to phenanthrene (e) and bioaccumulation
kinetic of phenanthrene (0) during the exposure period in B Lagoon (A), Scrubber (B),
SLR Downstream (C), EGSL04-07 (D), SLR Upstream (E), and Hospital Beach (F)
sediments. Note different scales for phenanthrene concentration.
A D
1.0 1.0 1.0 0.5
0.8 0.8 0.8 0.4 ~ '""" Li! :i Li! :;:
:t :i
:t :i c- 0.6 {'o. 0.6 c- 0.6 0.3 W .,- W .,-
:t Oll :t Oll C- Oll c- Oll
::i:. 0.4 0.4 2; ::i:. 0.4 0.2 2:: 0 Li! 0 r; Li! d- :r: d- :t
( c- c-0.2 0.2 0.2 0. 1
0.0 0.0 0.0 0.0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time (h) Time (h)
B E
1.0 10 1. 0 0.5
0.8 0.8 0.4
'""" '""" Li! :;: Li! :;: :t
:i :t
:i e:: 0.6 6 e:: 0.6 0.3 Li! .,- Li! .,-:t Oll :t Oll c- Oll c- Oll
::i:. 0.4 2; ::i:. 0.4 0.2 2; 0 Li! 0 Li! d- :t d- :t c- c-
0.2 0.2 0. 1
0.0 0 0.0 0.0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time (h) Time (h)
C F
1.0 0.10 1.0 0.5
0.8 0.08 0.8 0.4 Li! '"""Li! :i :t :;: :t
0.3 .,-:i ili 0.6 0.06 :;: e:: 0.6 .,- UJ
f\..... :t Oll;t Oll c- 0ll C- Oll
::i:. 0.4 0.04 ::1. ' ~:r: 0.4 0.2 -3
0 UJO UJ d- :r: ' :t c-o" c-
0.2 0.02 0.2 V 0. 1
0.0 0.00 0.0 0.0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time (h) Time (h)
126
Pyrene and 1-hydroxypyrene showed a similar behavior to phenanthrene and its
metabolite with a stable 1-0H-PYRIPYR ratio reaching a plateau after a few days (Fig.4).
A peak is often observed early in the process (Fig. 4A and C), but seems not related to the
concentration of pyrene in worms. 1-0H-PYRIPYR ratios ranged from 0.2 to 0.4 at steady
state with the exception of SLR Upstream (Fig. 4E) where the ratio stabilized around 0.1 as
previously observed for 9-0H-PHE/PHE.
Uptake rate constants (kt), metabolization rate constant (k2) and elimination rate
constant (k3) were calculated for individual OH-PAHs using Eq. 2 (Table 4). As previously
observed in Table 2 for uptake rate (ks) of the parent compound calculated with Eq.1 , the
uptake clearance rate constants (kt) of phenanthrene and pyrene calculated with Eq. 2 are
much lower in highly contaminated samples (averaged 9.3 x 10-4 for PHE and 3.4 x 10-4 for
PYR) than in low contaminated ones (averaged 0.10 for PHE and 0.18 for PYR).
Metabolization rate constants (k2) for phenanthrene ranged from 0.007 to 0.079 with
an exceptional high value of 0.35 for B Lagoon sediment. Rate k2 for pyrene ranged from
0.001 in EGSL04-13 to 0.14 in Hospital Beach.
The elimination rate (k3) of the phenanthrene metabolite ranged from 0.43 (in B
Lagoon where k2 is also high) to 0.0076 (in SLR Upstream where k2 was particularly low).
Rate k3 for pyrene ranged from 0.01 in SLR Downstream to 0.19 in Scow Grid.
127
Fig. 4. The ratio of I-hydroxypyrene to pyrene (. ) and bioaccumulation kinetic of pyrene
(0) during the exposure period in B Lagoon (A), Scrubber (B), SLR Downstream (C),
EGSL04-07 (D), SLR Upstream (E), and Hospital Beach (F) sediments. Note different
scales for pyrene concentration.
A D
1.0 1.0 0.5
0.8 0 .8 0.4 ~
~ :i ~ :i
~ 0.6 :i ~ 0.6 0 .3 :i
"7 "7 on >- on "- on "- on :i: 0.4 2 3 :i: 0.4 0.2 3 C? ~ C? / ~
>-"- "-0.2 0.2
~ 0. 1
0.0 0 0.0 0.0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time (h) Time (h)
B E
1.0 25 1.0 1.0
0.8 20 0.8 0.8 ~
~ ~ :i :i
~ 0.6 15 :i ~ 0.6 0.6 :i
"7 on "7 >- on c,- on "- on :I: 0.4 10 3 :i: 0.4 0.4 3 C? ~ C? ~
"- "-0.2 0.2 0.2
0.0 0.0 0 .0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time (h) Time (h)
C F
1.0 0.5 1.0 0.5
0.8 0.4 0.8 0.4
~ ~
~ ~
:i :i
~ 0.6 0.3 :i ~ 0.6 0.3 :i
"7 on "7 on "- on "- on :i: 0.4 0.2 3 :i: 0.4 ~ 0 .2 3 C? ~ C? ~
"- "-0.2 0. 1 0.2 0 . 1
0.0 0.0 0.0 0.0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700
Time (h) Time (h)
128
The relationship between the elimination rate constants (k3) and the metabolization
rate constants (k2) is iUustrated in Figure 5. The plots show that k3 increased with increasing
k2 for both metabolites in aU studied sediment samples with good correlation coefficients
(values of R2 between 0.52 and 0.95) with the exception of 9-0H-PHEN in low
contaminated sediments (R2 = 0.05). The best correlation coefficient is obtained for 1-0H-
PYR in highly contaminated sediments (R2 = 0.95).
129
Fig. 5. Plot of the elimination rate constants (k3) versus the metabolization rate constants
(k2) for 9-hydroxyphenanthrene in aIl (A), highly (B), and low contaminated sediments (C)
and for 1-hydroxypyrene in aIl CD), highly CE), and low contaminated sediments CF).
0.5
0.4
0.3 +
0.1
A
+
0.0 ~----"'-------,------,
~
:S
0.5
0.4
.::J:.M 0.2
0.1
o 0.4
B
0.0 ~--'-----r-------'
0.5
0.4
- 0.3 :S ~M 0.2
0.0 0 .2 0.4
c
+ R2 =0.05 .+ •
E. :2
0.5
0.4
0.3
0.00
0.5
0.4
- 0.3 :S ~M 0.2
0.1
o
0.05 0.10 0.15
E
R2 = 0.95
0 +-4.....".,-,-----,.-----,
0.00
0.5
0.4
0.05
F
0.10 0.15
. 0.3 :S ..>J:.M 0.2 R2 = 0.74
0.1~ o~,
o 0.05 0.1 0.15
130
DISCUSSION
Bioconcentration of P AH in worms
The accumulation of organic compounds by N. virens and L. variegatus is complex
and depends on the characteristics of the compound, its aging process in sediment, and the
organism toxicokinetic characteristics. When these complex interactions are sorted out, the
physicochemical properties of the compound are found to govern the accumulation of
P AHs. From the relationships with log Kow, we can say that the ks values of compounds
with log Kow greater than 7 will be extremely small. The relationship between log ks and
log Kow suggests that the desorption rate of the compound from the sediment is probably
important in governing its accumulation by the organism. The change in the kinetics with
increasing molecular lipophilicity expressed by log Kow has been previously observed for
Pontoporeia hoyi accumulation of P AHs from sediments (Landrum, 1989). Although the
slopes were similar (-0.62 and -0.64 for highly and low contaminated sediments
respectively), the displacement of the intercepts of the two lines (-0.29 and 2.4 for highly
and low contaminated sediments respectively) (Fig. 2 Top panel) is thought to reflect the
difference in bioavailability driven in part by differences in the total P AH content of the
sediment.
Previous studies have shown that fastest elimination rate is generally for compounds
with lower lipophilicity (Lotufo and Landrum, 2002). Our results are in contradiction with
these results. Indeed, the present study shows now significant differences (p > 0.05)
131
between compounds with low and high lipophilicity. The difference between these two
studies can mainly be due to the difference of studied species. Lotufo and Landrum (2002)
studied a freshwater amphipod (Diporeia spp), whereas we studied a freshwater oligochaete
(L. variegatus) and a marine polychaete (N. virens).
The propensity of individual P AHs to accumulate in worms can be estimated from
their bioaccumulation factor. In the present study, BAFs estimated from ks and ke values
were essentially similar to those estimated from P AH concentrations found in worm tissue
and sediment samples (Table 3). At both exposure levels (highly and low contaminated
sediments), BAFs decreased with increasing log Kow values for different PAHs (Table 3).
As explained in Barthe et al. (2008), the low BSAF observed in highly contaminated
samples can mainly be attributed to strong sorption to aluminum smelter residues weIl
characterized by Breedveld et al. (2007).
Biotransformation of P AHs in worms
The studies of PAH metabolization capabilities of L. variegatus are contradictory.
Indeed, sorne researchers (Harkey et al., 1994; Verragia Guerrero et al. , 2002) found no
PAH metabolites with L. variegatus whereas Leppanen and Kukkonen (2000), Lyytikainen
et al. (2007) and Schuler et al. (2003) found pyrene and benzo[a]pyrene metabolites in
organism tissues. The differences between these two groups of studies was explained by
Lyytikainen et al. (2007) who attributed these differences in results to differences in
experimental protocols. lndeed, the study by Harkey et al. (1994) was conducted at IOoe
and the one by Verrengia Guerrero et al. (2002) was performed during 48 h only under an
acute dosage regime, whereas studies by Leppanen and Kukkonen (2000), Lyytikainen et
132
al. (2007) and Schuler et al. (2003) were performed at 20°C during lOto 28 days. In the
present study, the bioaccumulation tests were performed at 4-6°C for 28 days and P AH
metabolites were observed in L. variegatus and in N virens for aIl sediments tested The
difference in temperature between the tests seems not to be the key factor to explain these
differences between results. Even time scale seems not to be a proper explanation as we
observed metabolites within 2-3 days after the beginning of the experiment. A part of the
explanation may reside in metabolite detection methods used by different research teams.
We observed that total pyrene in N virens tissues after a 28-day exposure was
constituted by 17% 1-hydroxypyrene and 83% pyrene which is in accordance with the
study of J0rgensen et al. (2005) who found 4% of I-hydroxypyrene and 17% of pyrene
toward total pyrene in N. virens gut tissue, which represents 23.5% of I-hydroxypyrene
compared to pyrene parent compound. According to this observation, we assumed that
metabolites present in worm were mainly due to metabolization of the parent P AH instead
of bioaccumulation from metabolites already present in low quantity in most sediment
samples. Moreover, as worms are able to biotransform I-hydroxypyrene and 9-
hydroxyphenanthrene (phase l metabolites) to phase II metabolites (Giessing and Lund,
2002; Giessing et al. , 2003; J0rgensen et al., 2005), we assume that the bioaccumulated
phase l metabolites would be biotransformed to phase II metabolite quickly and would not
be taken into account in the concentration measurements. The presence of 9-
hydroxyphenanthrene and I-hydroxypyrene, the phase I metabolites of phenanthrene and
pyrene, respectively, in the present study supports the capability of N. virens and L.
variegatus to metabolize 3- and 4-ring PAHs. This finding is in agreement with recent
133
Lyytikainen et al. (2007) study in which 1-hydroxypyrene was found In L. variegatus
tissues in sediment exposures with pyrene.
In the present study, the relative concentration of 9-hydroxyphenanthrene and 1-
hydroxypyrene compared to the concentration of phenanthrene and pyrene, respectively,
increased to a maximum, then decreased slightly in most case, and then reached the steady-
state level after 24 to 216 h. These results are in contrast with the recent results of
Lyytikainen et al. (2007) who found that the ratio of 1-hydroxypyrene/pyrene increased
linearly during the test and reached a value of ~ 1 after 15 days. Moreover, Lyytikainen et
al. (2007) observed that the concentration of I-hydroxypyrene was the highest at the
beginning of the test and decreased rapidly, whereas we found that the concentration of 9-
hydroxyphenanthrene and I-hydroxypyrene were the lowest at the beginning of the test and
increased rapidly to reach a steady-state level (data not shown). Reaching a steady-state
level means that elimination of the parent compound would be as efficient as
biotransformation in worm tissues. This observation is confirmed by the high correlation
coefficients obtained between the elimination rate constants (k3) and the metabolization rate
constants (k2) (Fig. 5).
CONCLUSION
The results of the present study showed that N. virens and L. variegatus accumulated
sediment-associated PAHs rapidly and achieved apparent steady-state within 7 days (168 h)
for aH studied sediment samples. In addition, the uptake clearance rate constant of P AHs
134
were lower for highly than for low contaminated sediments (10-2_10-6 and 10-1_10-4 g dry
sediment g-I wet organism h- I, respectively), whereas the elimination rate constant of PAHs
remained in same order of magnitude for both type of sediments. The relationship between
log ks and log Kow suggests that the desorption rate of the compound from the sediment is
probably important in governing the accumulation by the worms.
To our knowledge, the present study is the first one to report toxicokinetic data of
P AH biotransformation in metabolites. lndeed, other researchers have studied the
occurrence of phase 1 and phase II P AH metabolites in organisms during bioaccumulation
and toxicokinetic studies in polychaetes and oligochaetes (Christensen et al. , 2002;
J0fgensen et al. , 2005 ; Leppanen and Kukkonen, 2000; Lyytikainen et al. , 2007), but none
of them has studied accumulation, or metabolization, or elimination kinetics of these
metabolites. Our work opens a window in the study of the fate of P AH metabolites which
deserves more scientific attention from reseachers. Additional experiments are needed to
determine the possible input of microorganism-originating metabolites in worms versus the
metabolic formation rate of PAH metabolites by worms. lndeed, we assumed that the PAH
metabolites present in worms came mainly from the internaI biotransformation of
corresponding parent P AH in such a proportion that the bioaccumulated metabolites were
negligible, but it would be interesting to know this proportion in order to study their
bioaccumulation, biotransformation into phase II metabolites, and elimination kinetics.
135
ACKNOWLEDGEMENTS
This study was supported by a grant from the Canadian Research Chair in Molecular
Ecotoxicology (E.P.) and from the Natural Science and Engineering Research Council of
Canada (NSERC) - Collaborative Research Development Pro gram (CRD). Samples have
been provided by Alcan Ltée. This paper is a contribution of the Quebec-Ocean Network.
REFERENCES
Barthe M, Pelletier É. 2007. Comparing bulk extraction methods for chemically available
polycyclic aromatic hydrocarbons with bioaccumulation in worms. Environ Chem 4: 271-
283.
Barthe M, Pelletier É, Breedveld GD, Cornelissen G. 2008. Passive samplers versus
surfactant extraction for the evaluation of PAH availability in sediments with variable
levels of contamination. Chemosphere. 71: 1486-1493.
Breedveld GD, Pelletier É, St. Louis R, Cornelissen G. 2007. Sorption characteristics of
polycyclic aromatic hydrocarbons in aluminum smelter residues. Environ Sei Teehnol 41:
2542-2547.
Christensen M, Andersen 0 , Banta GT. 2002. Metabolism of pyrene by the polychaetes
Nereis diversieolor and Arenieola marina. Aquat Toxieol 58: 15-25.
DiToro D.M. , Zab ra CS, Hansen Dl, Berry Wl, Swartz RC, Cowan CE, Pavlou SP,
Allen HE, Thomas NA, Paquin PR, 1991. Technical basis for establishing sediment quality
136
criteria for nonionic orgamc chemicals using equilibrium partitioning. Environ Toxico.
Chem 10: 1541-1583.
Giessing AMB, Lund T. 2002 . Identification of 1-hydroxypyrene glucuronide in tissue
of marine polychaete Nereis diversicolor by liquid chromatography/ion trap multiple mass
spectrometry. Rapid Commun Mass Spectrom 16: 1521-1525.
Giessing AMB, Mayer LM, Forbes TL. 2003. 1-hydroxypyrene glucuronide as the major
aqueous pyrene metabolite in tissue and gut fluid from the marine deposit-feeding
polychaete Nereis diversicolor. Environ Toxicol Chem 22: 1107-1114.
Harkey GA, Landrum PF, Klaine Sl 1994. Comparison of who le-sediment, elutriate and
pore-water exposures for use in assessing sediment-associated organic contaminants in
bioassays. Environ Toxicol Chem 13: 1315-1329.
l0rgensen A, Giessing AB Rasmussen Ll, Andersen 0 , 2005 . Biotransformation of the
polycyclic aromatic hydrocarbon pyrene in the marine polychaetes Nereis virens. Environ
Toxicol Chem 24: 2796-2805 .
Kennish Ml 1997. Polycyclic aromatic hydrocarbons. In: Kennish, Ml (Eds.). Practical
Handbook of Estuarine and Marine Pollution. CRC Press, Boca Raton, FL, USA, pp. 141-
175.
Kukkonen lVK, Landrum PF, Mitra S, Gossiaux DC, Gunnarsson l , Weston D. 2004.
The role of desorption for describing the bioavailability of select polycyclic aromatic
hydrocarbon and polychlorinated biphenyl congeners for seven laboratory-spiked
sediments. Environ Toxicol Chem 23: 1842-1851.
137
Landrum PF. 1989. Bioavailability and toxicokinetics of polycyclic aromatic
hydrocarbons sorbed to sediments for the Amphipod Pontoporeia hoyi. Environ Sei
Technol23: 588-595.
Landrum PF, Eadie BJ, Faust WR. 1992a. Variation in the bioavailability of polycyclic
aromatic hydrocarbons to the amphipod Diporeia (spp.) with sediment aging. Environ
Toxicol Chem Il: 1197- 1208.
Landrum PF, Lee II H, Lydy Ml 1992b. Toxicokinetics in aquatic systems: Model
comparisons and use in hazard assessment. Environ Toxicol Chem Il: 1709- 1725.
Leppanen MT, Kukkonen JVK. 1998. Relative importance of ingested sediment and pore
water as bioaccumulation routes for pyrene to oligochaete (Lumbriculus variegatus,
Müller) . Environ Sei Technol32: 1503- 1508 .
Leppanen MT, Kukkonen JVK. 2000. Fate of sediment-associated pyrene and
benzo[a]pyrene in the freshwater oligochaetes Lumbriculus variegatus (Müller). Aquat
Toxicol49: 199-212.
Loonen H, Muir DCG, Parsons JR, Govers Al 1997. Bioaccumulation of
polychlorinated dibenzo-p-dioxins in sediment by oligochaetes: Influence of exposure
pathway and contact time. Environ Toxicol Chem 16: 1518- 1525.
Lotufo GR, Landrum PF. 2002. The influence of sediment and feeding on the elimination
of polycyclic aromatic hydrocarbons in the freshwater amphipod, Diporeia spp. Aquat
Toxicol58: 137-149.
Lydy MJ, Lasater JL, Landrum PF. 2000. Toxicokinetics of DDE and 2-chlorobiphenyl
in Chironomus tentans. Arch Environ Contam Toxicol38 : 163-168.
138
Lyytikainen M, Pehkonen S, Akkanen J, Leppanen MT, Kukkonen JVK. 2007.
Bioaccumulation and biotransformation of polycyclic aromatic hydrocarbons during
sediment tests with oligochaetes (Lumbriculus variegatus). Environ Toxicol Chem 26:
2660-2666.
Mackay D, Shiu WY, Ma KC, 1991. Illustrated Handbook of Physical-chemical
Properties and Environmental Fate of Organic Chemicals, Vol. 1. Lewis Publishers, Boca
Raton, FL, USA.
McElroy AE, Means JC. 1988. Factors affecting the bioavailability of
hexachlorobiphenyls to benthic organisms. In Adams WJ, Champman GA, Landis WG
(Eds.). Aquatic Toxicology and Hazard Assessment, Vol 10. STP 971. American Society
for Testing and Materials, Philadelphia, PA, USA, pp 149-158.
Pignatello Jl. 1990. Slowly reversible sorption of aliphatic hydrocarbons in soils. 1.
Formation ofresidual fractions. Environ Toxicol Chem 9: 1107- 1115.
Schuler LJ, Lydy Ml. 2001. Chemical and biological availability of sediment-sorbed
benzo[a]pyrene and hexachlorobiphenyl. Environ Toxicol Chem 20: 2014-2020.
Schuler LJ, Wheeler M, Bailer AJ, Lydy Ml. 2003. Toxicokinetics of sediment-sorbed
benzo[a]pyrene and hexachlorobiphenyl using the freshwater invertebrates Hyalella azteca,
Chironomus tentans, and Lumbriculus variegatus. Environ Toxicol Chem 22: 439-449.
Spacie A, Hamelink JL. 1983 Alternative models for describing the bioconcentration of
organics in fish. Environ Toxicol Chem 1: 309-320.
Spacie A, McCarty LS, Rang GM. 1995. Bioaccumulation and bioavailability in
multiphase systems. In Rand GM (Ed.). Fundamentals of Aquatic Toxicology: Effects,
139
Environmental Fate, and Risk Assessment. 2nd edition. Taylor & Francis, London, UK, pp.
493-521.
Varanasi U, Reichert WL, Stein JE, Brown DW, Sanborn HR. 1985. Bioavailability and
biotransformation of aromatic hydrocarbons in benthic organisms exposed to sediment
from an urban estuary. Environ Sei Teehnol19: 836-841.
Verrengia Guerrero NR, Taylor MG, Davies NA, Lawrence MAM, Edwards PA,
Simkiss K, Wider EA. 2002. Evidence of differences in the biotransformation of organic
contaminants in three species offreshwater invertebrates. Environ Pollut 117: 523-530.
Weston DP. 1990. Hydrocarbon bioaccumulation from contaminated sediment by
Abarenieola paeifiea, a deposit-feeding polychaete. Mar Biol 1 07: 159-169.
CONCLUSION GÉNÉRALE & PERSPECTIVES
141
Conclusions générales
Traditionnellement, les concentrations en hydrocarbures aromatiques polycycliques
(HAP) dans les sols et les sédiments ont été déterminées après des extractions solide/liquide
vigoureuses utilisant des solvants organiques, tels que le dichlorométhane ou
l ' hexane/acétone. Cependant, quand les contaminants organiques hydrophobes, tels que les
HAP, entrent dans les sols ou sédiments, ils subissent un certain nombre de processus de
perte ou de transport incluant la volatilisation et le lessivage, aussi bien que de la
dégradation chimique et biologique qui peuvent modifier leur comportement. De plus, les
processus de sorption peuvent permettre à ces composés chimiques de devenir plus
résistants à l 'extraction par des solvants ou d'être irréversiblement liés à la matrice interne
des sols ou sédiments. Le résultat de ce processus d' altération avec le temps est une
diminution de l' extractabilité des HAP. Ce phénomène a été appelé "effet de
vieillissement" et est relié à un déclin de leur disponibilité pour l' absorption par les
organismes benthiques et leur biodégradabilité par les microorganismes. Un effort
important en recherche a été récemment consacré au développement de méthodes
chimiques fiables afin de déterminer la partie biodisponible présente dans les sols et
sédiments.
Les travaux de cette thèse ont donc cherché à déterminer une ou plusieurs méthodes
chimiques dites douces ou sélectives permettant de caractériser les concentrations en HAP
biodisponibles dans des sédiments de niveaux de contamination et de propriétés
géochimiques variables mais également à établir les patrons temporels d 'accumulation des
142
HAP dans ces échantillons dans deux espèces de vers en identifiant les paramètres
toxicocinétiques décrivant l' ingestion, l'élimination et les facteurs de bioaccumulation.
Ainsi, trois séries d' expériences ont été effectuées. La première a permis de comparer
trois différentes méthodes d 'extraction solide/liquide (n-butanol (BuOH), hydroxypropyl-~
cyclodextrine (HPCD), et une solution de tensioactif de Brij700 (B700)) et la seconde a
comparé l' efficacité de deux échantillonneurs passifs, des bandes de polyoxyméthylène
(POM) et des tubes de polydimethylsiloxane (PDMS). Les résultats de ces deux études ont
été comparés avec les résultats de tests de bioaccumulation sur 28 jours effectués sur les
mêmes sédiments avec Nereis virens pour les sédiments marins et Lumbriculus variegatus
pour les sédiments lacustres. De plus, les résultats de la seconde étude ont été comparés
avec ceux obtenus lors de la première étude avec la solution aqueuse de B700. La troisième
expérience a permis de déterminer les patrons temporels d' accumulation des différents
HAP par les vers (N. virens et L. variegatus) dans les différents sédiments étudiés ainsi que
l' habilité des organismes à métaboliser le phénanthrène et le pyrène par la détermination
des hydroxy-HAP correspondants (9-hydroxyphénanthrène et 1-hydroxypyrène).
Les résultats obtenus lors de cette première étude montrent que des différences
importantes ont été observées aussi bien dans la quantité que dans les proportions des HAP
obtenues en utilisant différentes techniques pour déterminer la biodisponibilité des HAP
dans des sédiments. Bien que plusieurs recherches ont suggéré qu'une extraction dite douce
utilisant du BuOH pouvait être un moyen approprié pour la détermination des HAP
biodisponibles, nos résultats sont en désaccord avec les études précédentes aussi bien pour
les sédiments faiblement que fortement contaminés. Nos travaux montrent que le BuOH
143
extrayait plus que ce que les vers pouvaient bioaccumuler dans leurs tissus, et que le HPCD
sous-estimait la biodisponibilité des HAP spécialement les HAP de faible poids
moléculaire. D'autre part, l'utilisation du tensioactif B700 a montré une bonne capacité
prédictive pour les HAP dans les sédiments fortement contaminés. Les études en
laboratoire conduites avec du sédiment fraîchement additionné de HAP marqué peuvent
surestimer la biodisponibilité des contaminants comparée aux situations de terrain où des
temps d'équilibrage entre le sédiment et les contaminants persistants plus grands réduisent
cette biodisponibilité (Conrad et al., 2002; Leppanen et Kukkonen 2000a). De plus, la
plupart de ces études déterminaient la biodisponibilité en mesurant la dégradation
microbienne, une approche qui peut ne pas être comparable avec la bioaccumulation des
HAP par des vers puisque les vers ne sont pas des échantillonneurs passifs et peuvent
modifier le profil du HAP dans leurs tissus.
Puisque l' évaluation du risque des sols contaminés est actuellement surestimée par
l'utilisation d'analyses chimiques qui s'en remettent à une extraction vigoureuse initiale, il
est essentiel d'inclure la détermination de la bioaccumulation dans les décisions de
réglementation. L'utilisation de l'ingestion par les vers et la biodégradation sont encore les
meilleurs outils pour estimer la biodisponibilité des HAP, mais les méthodes biologiques
sont longues et coûteuses. Nos résultats illustrent la difficulté à trouver une méthode
chimique adéquate pour prédire la biodisponibilité des HAP, particulièrement parce que les
concentrations en HAP et le processus de séquestration jouent un rôle déterminant dans la
qualité des résultats et ceci en fonction de paramètres environnementaux et historiques le
plus souvent inconnus de l'utilisateur. Puisque le B700 est peu coûteux et que les solutions
144
sont simples à préparer, une procédure d'extraction utilisant ce tensioactif de haut poids
moléculaire se révèle être une méthode utile particulièrement pour les sédiments présentant
des niveaux en HAP >2 Ilg g- I et avec des log Kow <5,8.
Les résultats de notre seconde étude montrent que les deux échantillonneurs passifs
agissent de manière différente par rapport aux mêmes échantillons de sédiment. En effet, le
PDMS surestime la disponibilité des HAP dans les sédiments étudiés (faiblement et
fortement contaminés), alors que le POM produit des résultats similaires à l'extraction au
B700. Quoi qu'il en soit, le POM et le B700 prédisent mal la disponibilité des HAP dans
les sédiments faiblement contaminés où les facteurs biologiques (matière organique
digestible) deviennent importants. Par contre, la biodisponibilité des HAP totaux a été
prédite correctement par le POM et le B700 dans les sédiments fortement contaminés par
des alumineries. Un examen plus détaillé des résultats obtenus pour les HAP individuels
indique que les deux techniques surestiment la disponibilité des grosses molécules avec un
log Kow >6 suggérant un mécanisme biologique limitant l'incorporation des plus grosses
molécules de HAP ce qui semble être relié à la taille moléculaire des composés ou à leur
encombrement stérique; un mécanisme non directement relié à la liposolubilité de la
molécule.
Les résultats de la troisième étude ont montré que Nereis virens et Lumbriculus
variegatus accumulaient rapidement les HAP associés au sédiment et atteignaient un état-
stable apparent en 7 jours (168 h) pour tous les sédiments étudiés. De plus, les constantes
du taux d'ingestion des HAP sont plus faibles pour les sédiments fortement contaminés que
pour les faiblement contaminés (10-2_10-6 et 10-1_10-4 g sédiment sec g-I organisme humide
145
h- l, respectivement), alors que la constante du taux d'élimination des HAP sont du même
ordre de grandeur pour les deux types de sédiments. La relation négative entre log ks et log
Kow suggère que le taux de désorption du composé à partir du sédiment est probablement
important pour gouverner l'accumulation par les vers.
A notre connaissance, l'étude présente est la première à rapporter des données de
toxicocinétiques de biotransformation de HAP en métabolite. En effet, les autres chercheurs
ont étudiés l' apparition des métabolites de phase 1 et II dans les organismes pendant les
études de bioaccumulation et de toxicocinétiques chez les polychètes et oligochètes
(Christensen et al. , 2002; J0rgensen et al. , 2005; Leppanen et Kukkonen, 2000b;
Lyytikainen et al. , 2007), mais aucun de ces chercheurs n' avaient étudié les cinétiques
d ' accumulation, ou de métabolisation ou d'élimination de ces métabolites. Nos travaux
ouvrent une porte dans l' étude du devenir des métabolites de HAP qui nécessite une
attention scientifique plus importante de la part des chercheurs. Des expériences
additiOlmelles sont nécessaires afin de déterminer l' apport possible de métabolites
provenant de microorganismes dans les vers par rapport au taux de transformation des HAP
par les vers eux-mêmes. En effet, nous avons présumé que les métabolites présents dans les
vers provenaient principalement d'une biotransformation interne des HAP parents
correspondants dans une telle proportion que les concentrations de métabolites
bioaccumulés étaient négligeables, mais il serait intéressant de connaître cette proportion
afin d 'étudier leurs cinétiques de bioaccumulation, de biotransformation en métabolites de
phase II, et d' élimination.
146
Conclusion
La réalisation de ces travaux nous aura permis d'atteindre l'ensemble des objectifs
fixés. Nous avons pu mettre au point deux méthodes d ' extraction sélective permettant de
déterminer les HAP totaux biodisponibles dans les sédiments fortement contaminés par des
alumineries et plus précisément les HAP avec un log Kow <5,8. De plus, l 'étude de la
bioaccumulation nous a montré que la composition et le niveau de contamination jouaient
un rôle sur la biodisponibilité des HAP. On a pu voir que les vers bioaccumulaient, en
proportion, plus de HAP dans les sédiments faiblement contaminés que dans les fortement
contaminés. Cette étude nous a aussi permis de déterminer que les divers paramètres
toxicocinétiques étudiés (ingestion, élimination, facteurs de bioaccumulation) variaient
d 'un sédiment à l' autre et d'un HAP à l'autre.
Tout au long de ce travail, nous avons parlé de biodisponibilité en général sans
prendre en compte les définitions établies par Semple et al. (2004) sur la biodisponibilité et
la bioaccessibilité car le but de ce travail n'était pas d'entrer dans le débat à propos des
définitions pratiques de ces deux termes. Nous allons maintenant tenter de faire un lien
entre ces définitions et nos travaux. Récemment ter Laak et al. (2006) ont scindé en deux
les différentes méthodes d' extraction selon les critères de bioaccessibilité et de
biodisponibilité. Premièrement, les méthodes qui extraient la fraction faiblement liée, dont
fait partie le B700, et qui est selon ter Laak et al. (2006) la fraction bioaccessible, et
deuxièmement les méthodes qui extraient les concentrations dissoutes libres présentes dans
les eaux porales, ce qui inclus les échantillonneurs passifs tels que le POM et le PDMS, et
147
qui est appelée la fraction biodisponible. Selon les définitions de Semple et al. (2004), la
partie bioaccessible est en théorie plus importante que la partie biodisponible puisque, selon
ces mêmes auteurs, la bioaccessibilité englobe ce qui est réellement biodisponible et ce qui
est potentiellement biodisponible. Théoriquement, si on se fie à ces définitions, les
concentrations extraites par le B700 devraient être plus importantes que celles extraites par
le POM ou les vers puisque ces deux dernières méthodes déterminent la partie
biodisponible d'un contaminant. Or nos résultats montrent que ce n'est pas le cas. En effet,
comme on a pu le voir dans le chapitre 2, les résultats obtenus avec le B700 sont similaires
à ceux du POM c'est à dire une bonne estimation de la biodisponibilité des HAP de log Kow
<6 dans des sédiments fortement contaminés par des rejets d'alumineries . Ceci peut
s'expliquer par le fait que l' extraction des concentrations biodisponibles dans les différents
sédiments ne se fait pas instantanément. En effet, celle-ci dure 16 h pour le B700 et 21
jours pour le POM. En fait, les deux méthodes chimiques, ainsi que la méthode biologique
(bioaccumulation par les vers), extraient la partie bioaccessible du contaminant car, selon la
définition de Semple et al. (2004), la fraction biodisponible est la fraction librement
disponible à un moment donné et la fraction bioaccessible est la fraction qui est disponible
à ce moment donné et celle qui sera disponible plus tard.
Perspectives
De nombreux points demandent encore à être explorés et concernent en premier lieu
l'aspect purement chimique. En effet, au cours de ces travaux, on a pu voir que les
méthodes que nous avons proposées comme étant des méthodes adéquates pour prédire les
concentrations biodisponibilites des HAP (B700 et POM) avaient des comportements
148
différents par rapport aux sédiments faiblement et fortement contaminés. Donc, le premier
point qui pourrait être à développer est la détermination de la "limite de détection" de ces
méthodes afin de définir la concentration en HAP totaux à partir de laquelle les
concentrations extraites avec les méthodes chimiques deviennent en accord avec celles
biodisponibles pour les organismes. Lors de nos travaux, le sédiment avec la concentration
la plus élevée parmi les échantillons faiblement contaminés était la station EGSL04-07 avec
une concentration totale en HAP de 1, 1 ~g g-l et celui avec la concentration la moins élevée
parmi les échantillons fortement contaminés était la station SLR Downstream avec une
concentration totale en HAP de 25,5 ~g g-l. On peut voir qu ' il y a un facteur 23 entre les
deux sédiments. Le but de cette étude serait donc de combler le trou entre ces deux
échantillons en trouvant des échantillons naturels avec des concentrations en HAP totaux
comprises entre 1,1 et 25,5 ~g g-l.
Deuxièmement, afin d' étendre la connaissance et les applications du POM et du B700
pour des composés organiques autres que les HAP, d 'autres expériences pourraient être
effectuées afin d 'explorer leur performance à mesurer la fraction disponible de composés
organiques tels que les BPC. Le POM a une structure rigide et une excellente résistance
mécanique, ce qui fait de lui un échantillonneur potentiel pour des mesures in situ de la
concentration dans l'eau porale. Quoi qu'il en soit, il y a encore quelques considérations à
aborder par rapport au déploiement sur le terrain. Premièrement, il est difficile de contrôler
la température sur le terrain et les variations de la température peuvent affecter l' état
d 'équilibre et les taux d' extraction des HAP par le POM. Ainsi il est essentiel de déterminer
la déviation des performances du POM par rapport au calibrage fait en laboratoire causée
149
par les fluctuations de température sur le terrain. Ceci peut être accompli en effectuant des
études sur les isothermes de sorption et sur la cinétique d'extraction en laboratoire à
différentes températures contrôlées. Deuxièmement, la complexité de l' hydrodynamisme
rencontré sur le terrain peut aussi influencer les taux d'échantillonnage du POM par rapport
aux composés cibles; ainsi plusieurs tests sont suggérés afin d' examiner les effets de
différents débits d'eau. De plus, d'autres paramètres importants tels que l'épaisseur de la
membrane, l'aire de la surface de contact doivent aussi être évalués soigneusement afin
d'optimiser les performances du POM dans ses applications sur le terrain.
Certains aspects biologiques peuvent aussi être explorés. En premIer lieu,
Lumbriculus variegatus n'est généralement pas testé à des températures aussi faibles (4 à
6°C) mais à une température d'environ 20°C. Deux possibilités s'offrent à nous, la
première étant de recommencer les expériences de bioaccumulation des sédiments
dulcicoles à une température de 20°C et la deuxième étant de trouver un autre organisme
qui peut être utilisé à une température proche de celle que nous avons étudiée. Lors de nos
tests de bioaccumulation, nous avions choisi d' avoir le moins de différence possible entre
nos tests avec les sédiments marins et les sédiments lacustres. C'est pour cela que les tests
avec les sédiments lacustres ont eu lieu à 4-6°C. Afin de continuer dans ce processus, il
serait préférable d'utiliser des organismes qui peuvent être testés à ces faibles températures.
Deux espèces d'oligochaetes d'eau douce peuvent être utilisées, Limnodrilus hoffmeisteri et
Tubifex tubifex ainsi qu' une espèce d'amphipode, Diporeia spp .. De plus, une étude a déjà
utilisé L variegatus à une température de 10°C (Kraaij et al., 2002).
150
De plus, lors des études de bioaccumulation seule une espèce marine (Nereis virens)
et une espèce dulcicole (Lumbriculus variegatus) ont été utilisées. Bien que ces deux
espèces soient recommandées par l'Agence de Protection de l'Environnement Américaine
(US EPA) pour les tests de bioaccumulation dans les sédiments, d'autres espèces peuvent et
sont utilisées pour ces tests. En effet, de nombreux chercheurs utilisent d' autres polychaetes
(Nereis diversicolor, Arenicola marina) (Cornelissen et al. , 2006; Kaag et al. , 1997), des
bivalves (Mytilus edulis , Macoma balthica) (Kaag et al., 1997), des gastéropodes (Hinia
reticulata) (Cornelissen et al., 2006), des amphipodes (Diporeia spp.) (Landrum et al. ,
2007) ou bien la biodégradation par des bactéries (Liste et Alexander, 2002) pour
déterminer la biodisponibilité des contaminants étudiés. Comme ces organismes ont déjà
été utilisés, il nous serait possible d'effectuer des tests de bioaccumulation avec ces
différents organismes afin de déterminer si les deux méthodes chimiques peuvent prédire la
biodisponibilité des HAP seulement pour les deux sortes de vers ou alors pour d'autres
sortes d' espèces.
Finalement, nous avons étudié seulement deux métabolites de HAP, le 9-
hydroxyphénanthrène et le I-hydroxypyrène. Une autre perspective serait de déterminer des
métabolites d' autres HAP tels que le benzo[a]pyrène, le chrysène ou le naphtalène ou
encore de déterminer d'autres métabolites du phénanthrène et du pyrène. De plus, il serait
aussi bon de déterminer avec précision si les courbes de cinétique de métabolisation
obtenues lors de l' étude de bioaccumulation représentent bien la métabolisation ou si elles
sont des courbes de bioaccumulation de métabolites ou encore si elles représentent un
mélange de bioaccumulation des métabolites et de métabolisation. Pour ce faire,
151
l'utilisation de métabolites marqués au C-13 serait un bon moyen pour déterminer quelle
hypothèse est la bonne et si c'est la troisième, cela permettrait de déterminer en quelle
proportion.
BIBLIOGRAPHIE
153
Akkanen, 1.; Penttinen, S.; Haitzer, M.; Kukkonen, J.V.K. 2001. Bioavailability of
atrazine, pyrene and benzo[a]pyrene in European river waters. Chemosphere. 45: 453-
462.
Alexander, M. 2000. Aging, bioavailability, and overestimation of risk from
environmental pollutants. Environ. Sei. Technol. 34: 4259-4265.
Alexander, R.R.; Alexander, M. 2000. Bioavailability of genotoxic compounds in soils.
Environ. Sci. Technol. 34: 1589-1593.
Barthe, M. 2002. Étude de la Séquestration Chimique des Hydrocarbures Aromatiques
Polycycliques (HAP) par Extraction Sélective de Sédiments Lacustres et Marins. M. Sc.
Océanographie, Université du Québec à Rimouski. Mémoire de maîtrise. 200 pp.
Bowerman, M.E. 2004. Long-term Survival of Early Stage Fish Exposed to Polycyclic
Aromatic Hydrocarbon-Contaminated Sediment. M. Sc. in Biology, Queen ' s
University. Mémoire de maîtrise. 131 pp.
Chang, M.-C.; Huang, C.-R.; Shu, H.-Y. 2000. Effects of surfactants on extraction of
phenanthrene in spiked sand. Chemosphere. 41: 1295-1300.
Christensen M, Andersen 0, Banta GT. 2002. Metabolism of pyrene by the polychaetes
Nereis diversicolor and Arenicola marina. Aquat Toxicol 58: 15-25.
Christensen, E.R.; Bzdusek, P.A. 2005 . PAHs in sediments of the Black River and the
Ashtabula River, Ohio: source apportionment by factor analysis. Water Res. 39: 511 -
524.
154
Chung, N.; Alexander, M. 1999a. Effect of concentration on sequestration and
bioavailability of two polycyclic aromatic hydrocarbons. Environ. Sei. Teehnol. 33:
3605-3608 .
Chung, N.; Alexander, M. 1999b. Relationship between nanoporosity and other
properties ofsoil. Soil Science. 164: 726-730.
Chung, N.; Alexander, M. 2002. Effect of soil properties on bioavailability and
extractability of phenanthrene and atrazine sequestered in soil. Chemosphere. 48: 109-
115.
Conrad, A.U.; Comber, A.D.; Simkiss, K. 2002. Pyrene bioavailability; effect of
sediment-chemical contact time on routes of uptake in an oligochaete worm,
Chemosphere, 49: 447-454.
Cornelissen, G.; Ri gterink, H.; Ferdinandy, M.M.A.; van Noort, P.C.M. 1998. Rapidly
desorbing fractions of P AHs in contaminated sediments as a predictor of the extent of
bioremediation. Environ. Sci. Teehnol. 32: 966-970.
Cornelissen, G.; Breedveld, G.D.; Naes, K.; Oen, A.M.P.; Ruus, A. 2006.
Bioaccumulation of native polycyclic aromatic hydrocarbons from sediment by a
polychaetes and a gastropod: freely dissolved concentrations and activated carbon
amendment. Environ. Toxieol. Chem. 25: 2349-2355.
Cornelissen, G.; Pettersen, A.; Broman, D.; Mayer, P. ; Breedveld, G.D. in press. Field
testing of equilibrium passive sampi ers to determine free ly dissolved native polycyclic
aromatic hydrocarbon concentrations. Environ. Toxieol. Chem.
155
Cuypers, c.; Grotenhuis, T ; Joziase, l ; Rulkens, W. 2000. Rapid persulfate oxidation
predicts PAH bioavailability in soils and sediments. Environ. Sei. Technol. 34: 2057-
2063.
Cuypers, c.; Pancras, T; Grotenhuis, T; Rulkens, W. 2002. The estimation of PAH
bioavailability in contaminated sediment using hydroxypropyl-~-cyclodextrin and triton
X-100 extraction techniques. Chemosphere. 46: 1235-1245.
Dabestani, R. ; Ivanov, 1. 1999. A compilation of physical spectroscopic and
photophysical properties of polycyclic aromatic hydrocarbons. Photochemistry and
Photobiology. 70: 10-34.
Decker, G.C.; Bruce, W.N.; Bigger, lH. 1965. The accumulation and dissipation of
residues resulting from the use ofaldrin in soils. J Econ. Entomol. 58: 266-271.
Dickhut, R.M. ; Canuel, E.A.; Gustafson, K.E.; Liu, K.; Arzayus, K.M.; Walker, S.E. ;
Edgecombe, G.; Gaylor, M.O.; MacDonald, E.H. 2000. Automotive sources of
carcinogenic polycyclic aromatic hydrocarbons associated with particulate matter in the
Chesapeake Bay region. Environ. Sei. Technol. 34: 4635-4640.
Dungavell, J.A. 2005. The EfJects of Temperature on Po!ycyclic Aromatic Hydrocarbon
(PAH) Exposure and Metabolism by Fish. M. Sc. in Biology, Queen's University.
Mémoire de maîtrise. 108 pp.
Edwards, c.A., Beck, S.D.; Lichtenstein, E.P. 1957. Bioassay of aldrin and lindane in
soil. J Econ. Entomo!. 50: 622-626.
Erickson, D.C. ; Loehr, R.C.; Neuhauser, E.F. 1993. PAH loss during bioremediation of
manufactured gas plant site soils. Water Res. 27: 911-919.
156
Fava, F. ; Berselli, S.; Conte, P. ; Piccolo, A.; Marchetti, L. 2004. Effects of humic
substances and soya lecithin on the aerobic bioremediation of a soil historically
contaminated by polycyclic aromatic hydrocarbons (PAHs). Bioteehnol. Bioeng. 88 :
214-233.
Fetzer, J.e. 2000. The Chemistry and Analysis of the Large Polyeyclie Aromatie
Hydroearbons. Wiley, New York. 304 pp.
Ghosh, o.; Zimmerman, J.R.; Luthy, R.G. 2003. PCB and PAH speciation among
particle types in contaminated harbor sediments and effects on PAH bioavailability.
Environ. Sei. Teehnol. 37: 2209-2217.
Gilbert, W.S. ; Lewis, C.E. 1982. Residues in soil, pasture and grazing dairy cattle after
the incorporation of dieldrin and heptach10r into soil before sowing. Austr. J Exp.
Agrie. Anim. Husb. 22: 106-115.
Gulnick, J.D. 2000. Factors Injlueneing the Mierobial Mineralization of Polyeyclie
Aromatie Hydroearbons in Coastal Marine Sediments. PhD in Coastal Oceanography.
State University of New York at Stony Brook. Thèse de doctorat. 188 pp.
Hawthorne, S.B. ; Grabanski, C.B. 2000. Correlating selctive supercritical fluid
extraction with bioremediation behavior of P AHs in a field treatment plot. Environ. Sei.
Teehnol. 34: 4103-4110.
Herrchen, M. ; Debus, R.; Pramanik-Strehlow, R. 1997. Bioavailability as a Key
Property in Terrestrial Eeotoxieity Assessment and Evaluation, Fraunhofer IRB Verlag,
Stuttgart, Germany.
157
J0rgensen, A; Giessing, AM.B.; Rasmussen, L.J.; Andersen, O. 2005 .
Biotransformation of the polycyclic aromatic hydrocarbon pyrene ln the manne
polychaetes Nereis virens. Environ. Toxicol. Chem. 24: 2796-2805.
Kaag, N.H.B.M.; Foekema, E.M.; Scholten, M.C.Th.; van Straalen, N.M. 1997.
Comparison of contaminant accumulation in three species of marine invertebrates with
different feeding habits. Environ. Toxicol. Chem. 16: 837-842.
Klaasen, C.D. 1986. Distribution, excretion, and absorption of toxicants. Dans Casarett
and Doull's Toxicology: The basic Science of Poisons, 3rd Edition, Klaasen, C.D. ;
Amdur, M.O. ; Doull, J. (Eds), Macmillan, New York, 33-63.
Korschgen, L.1. 1971. Disappearance and persistence of aldrin after five annual
applications. J Wildlife Manage. 35: 494-500.
Kottler, B.D.; Alexander, M. 200l. Relationship of properties of polycyclic aromatic
hydrocarbons to sequestration in soil. Environ. Pollut. 113: 293-298.
Kraaij , R. ; Seinen, W.; ToUs, 1.; Cornelissen, G.; Belfroid, AC. 2002. Direct evidence
of sequestration in sediments affecting the bioavailability of hydrophobic orgamc
chemicals to benthic deposit-feeders. Environ. Sci. Technol. 36: 3525-3529.
Kramer, B.K.; Ryan, P.B. 2000. Soxhlet and microwave extraction in determining the
bioaccessibility of pesticides from soil and model solids. Proceedings of the 2000
Conference on Hazardous Waste Research: Environmental Challenges and Solutions 10
Resource Development, Production, and Use. 196-210.
158
Krauss, M.; Wilcke, W. 2001. Biomimetic extraction ofPAHs and PCBs from soil with
octadecyl-modified silica disks to predict their availability to earthworms. Environ. Sei.
Teehnol. 35: 3931 -3935.
Kukkonen, lV.K.; Landrum, P.F.; Mitra, S. ; Gossiaux, D.C. ; Gunnarsson, l ; Weston,
D. 2003. Sediment characteristics affecting desorption kinetics of select PAH and PCB
congeners for seven laboratory spiked sediments. Environ. Sei. Teehnol. 37: 4656-4663 .
Landrum, P.F. ; Robinson, S.D. ; Gossiaux, D.C.; You, l; Lydy, M.J. ; Mitra, S. ; Ten
Huslscher, T.E.M. 2007. Predicting bioavailability of sediment-associated organic
contaminants for Diporeia spp. and oligochaetes. Environ. Sei. Teehnol. 41: 6442-6447.
Law, R. J. ; Kelly, C. A. ; Baker, K. L.; Langford, K. H. ; Bartlett, T. 2002. Polycyclic
aromatic hydrocarbons in sediments, mussels and crustace a around a former gasworks
site in Shoreham-by-Sea, UK. Mar. Pollut. Bull. 44: 903-911.
Leppanen, M.T.; KUkkonen, J.V.K. 2000a. Effect of sediment-chemical contact time on
availability of sediment-associated pyrene and benzo[a]pyrene to oligochaete worms
and semi-permeable membrane devices. Aquat. Toxieol. 49: 227-241.
Leppanen, M.T. ; Kukkonen, J.V.K. 2000b. Fate of sediment-associated pyrene and
benzo[a]pyrene in the freshwater oligochaetes Lumbrieulus variegatus (Müller). Aquat.
Toxieol. 49 : 199-212.
Librando, V.; Hutzinger, O. ; Tringali, G.; Aresta, M. 2004. Supercritical tluid
extraction or polycyclic aromatic hydrocarbons from marine sediments and soils
samples. Chemosphere. 54: 1189-1197.
159
Lichtenstein, E.P. ; De Pew, L.1.; Eshbaugh, E.L.; Sleesman, lP. 1960. Persistence of
DDT, aldrin and lindane in sorne Midwestern soils. J Econ. Entomol. 53: 136-142.
Lichtenstein, E.P.; Schulz, K.R. 1965. Residues of aldrin and heptachlor in soils and
their translocation into various crops. J Agric. Food Chem. 13: 57-63.
Lima, A.L. ; Farrington, J.W.; Reddy, C.M. 2005 . Combustion-derived polycyclic
aromatic hydrocarbons in the environrnent - a review. Environ. Forensics. 6: 109-131.
Liste, H.H.; Alexander M. 2002. Butanol extraction to predict bioavailability of PAHs
in soil. Chemosphere. 46: 1011-1017.
Liu, Z. ; Laha, S.; Luthy, R.G. 1991. Surfactant solubilization of polycyclic aromatic
hydrocarbon compounds in soil-water suspensions. Water Sei. Teehnol. 23: 475-485.
Loibner, A.P.; Holzer, M.; Gartner, M.; Szolar, O.H.1.; Braun, R .. 2000. The use of
sequential supercritical fluid extraction for bioavailability investigations of P AH in soil.
Bodenkultur. 51: 279-287.
Lu, X. 2003 . Bioavailability and Bioaccumulation of Sediment-Associated, Desorption-
Resistant Fraction of Polyeyclie Aromatie Hydroearbon Contaminants. PhD in
Chemical Engineering. Louisiana State University and Agricultural & Mechanical
College. Thèse de doctorat. 162 pp.
Luthy, R.G.; Aiken, G.R.; Brusseau, M.L.; Cunningham, S.D.; Gschwend, P.M.;
Pignatello, J.1.; Reinhard, M.; Traine, S.1.; Weber, W.1. , Jr.; Westall, le. 1997.
Sequestration of hydrophobie organic contaminants by geosorbents. Environ. Sei.
Technol. 31: 3341-3347.
160
Lyytikainen M, Pehkonen S, Akkanen J, Leppanen MT, Kukkonen JVK. 2007.
Bioaccumulation and biotransformation of polycyclic aromatic hydrocarbons during
sediment tests with oligochaetes (Lumbrieulus variegatus). Environ Toxieol Chem 26:
2660-2666.
Mackay, D.; Shiu, W.Y.; Ma, K.C. 1991. Illustrated Handbook of Physieal-ehemieal
Properties and Environmental Fate of Organie Chemieals, Vol. 1. Lewis Publishers,
Boca Raton, FL, USA
Macleod, C.J.A; Semple, K.T. 2000. Influence of contact time on extractability and
degradation of pyrene in soils. Environ. Sei. Teehnol. 34: 4952-4957.
McElroy, AE.; Farrington, J. W.; Teal, J.M. 1989. Bioavailability of polycyclic
aromatic hydrocarbons in the aquatic environment. Dans Metabolism of Polyeyclie
Aromatie Hydroearbons in the Aquatie Environment, Varanasi, U. (Ed.), CRC Press,
Boca Raton, FL, 1-39.
Meier, J.R.; Chang, L.W.; Jacobs, S. ; Torsella, J.; Meckes, M.C.; Smith, M.K. 1997.
Use of plant and earthworm bioassays to evaluate remediation of soil from a site
contaminated with polychlorinated biphenyls. Environ. Toxieol. Chem. 16: 928-938.
Morrison, D.E. ; Robertson, B.K. ; Alexander, M. 2000. Bioavailability to earthworms of
aged DDT, DDE, DDD and dieldrin in soil. Environ. Sei. Teehnol. 34: 709-713.
Nam, K. ; Alexander, M. 1998. Role of nanoporosity and hydrophobicity In
sequestration and bioavailability: tests with model solids. Environ. Sei. Teehnol. 32: 71-
74.
161
Nam, K. ; Chung, N.; Alexander, M. 1998. Relationship between organic matter content
ofsoil and the sequestration of phenanthrene. Environ. Sei. Teehnol. 32: 3785-3788.
Oleszczuk, P.; Baran, S. 2004. The concentration of mild-extracted polycyclic aromatic
hydrocarbons in sewage sludges. J Environ. Sei. Health Part A Toxie-Hazard. 39:
2799-2815.
Onsager, J.A. ; Rusk, H. W.; Butler, L.I. 1970. Residues of aldrin, dieldrin, chlordane
and DDT in soil and sugar beets. J Eeon. Entomo!. 63: 1143-1146.
Oros, D.R. ; Ross, J.R.M. 2004. PolycYclic aromatic hydrocarbons in San Francisco
Estuary sediments. Mar. Chem. 86: 169-184.
Peijnenburg, W. ; Baerselman, R.; de Groot, A.; Jager, T.; Leenders, J.D. ; Posthuma, L.;
Van Veen, R.P.M. 2000. Quantification of metal bioavailability for lettuce (Laetuea
sativa L.) in field soils. Areh. Environ. Contam. Toxieo!. 39: 420-430.
Pignatello, J.1.; Xing, B. 1996. Mechanisms of slow sorption of organic chemicals to
natural particles. Environ. Sei. Teehnol. 30: 1-11.
Reid, B.1. ; Stokes, J.D.; Jones, K.C.; Semple, K.T. 2000. Nonexhaustive cYclodextrin-
based extraction technique for the evaluation of P AH bioavailability. Environ. Sei.
Teehnol. 34: 3174-3179.
Richardson, B.1.; Zheng, G.1.; Tse, E.S.C.; De Luca-Abbott, S.B.; Siu, S.Y.M. ; Lam,
P.K.S. 2003. A comparison of polycyclic aromatic hydrocarbon and petroleum
hydrocarbon uptake by mussels (Perna viridis) and semi-permeable membrane devices
(SPMDs) in Hong Kong coastal waters. Environ. Pollut. 122: 223-2227 .
162
Ruby, M.V.; Davis, A.; Schoof, R.; Eberle, S.; Sellstone, C.M. 1996. Estimation oflead
and arsenic bioavailability using a physiologically based extraction test. Environ. Sei.
Teehnol. 30: 422-430.
Rust, A.l; Burgess, R.M.; McElroy, A.E.; Cantwell , M.G. ; Brownawell, BJ. 2004a.
Influence of soot carbon on the bioaccumulation of sediment-bound polycyclic aromatic
hydrocarbons by marine benthic invertebrates: an interspecies comparison. Environ.
Toxieol. Chem. 23: 2594-2603.
Rust, A.l ; Burgess, R.M.; McElroy, A.E.; Cantwell , M.G. ; Brownawell, B.l. 2004b.
Role of source matrix in the bioavailability of polycyclic aromatic hydrocarbons to
deposit-feeding benthic invertebrates. Environ. Toxieol. Chem. 23: 2604-2610.
Semple, K.T. ; Doick, KJ. ; Jones, K.C.; Burauel, P.; Craven, A.; Harms, H. 2004.
Defining bioavailability and bioaccessibility of contaminated soil and sediment is
complicated. Environ. Sei. Teehnol. A-Pages, 38: 228A-231A.
Shor, L.M.; Rockne, KJ.; Taghon, G.L.; Young, L.Y. ; Kosson, D.S. 2003 . Desorption
kinetics for field-aged polycyclic aromatic hydrocarbons from sediments. Environ. Sei.
Teehnol. 37: 1535-1544.
Spacie, A. ; McCarty, L.S.; Rand, G.M. 1995. Bioaccumulation and bioavailability in
multiphase systems. Dans Fundamentals of Aquatic Toxieology: Effeets, Environmental
Fate and Risk Assessment, 2nd Edition, Rand, G.M. (Ed), Tyler & Francis, London,
493-522.
Swindell, A.L. ; Reid, BJ. 2006. Comparison of selected non-exhaustive extraction
techniques to assess PAH availability in dissimilar soils. Chemosphere . 62: 1126-1134.
163
Szolar, O.H.J.; Rost, H.; Braun, R.; Loibner, A.P. 2001. Assessment of (bio)availability
of PAHs in soil using sequential supercritical fluid extraction (SSFE). Dans
Kalogerakis, N. (Ed.), Proeeedings of the First European Bioremediation Conference .
Technical University, Chania, Crete, Grèce, 63-66.
Szolar, O.H.J.; Rost, H.; Hirmann, D.; Hasinger, M.; Braun, R. ; Loibner, A.P. 2004.
Sequential supercritical fluid extraction (SSFE) for estimating the availability of high
molecular weight polycyclic aromatic hydrocarbons in historically polluted soils. J
Environ. QuaI. 33: 80-88.
Talley, J.W. ; Ghosh, U; Tucker, S.G.; Furey, lS.; Luthy, R.G. 2002. Particle-scale
understanding of the bioavailability of PAH in sediment. Environ. Sei. Teehnol. 36:
477-483.
Tang, l; Liste, H.H.; Alexander, M. 2002. Chemical assays of availability to
earthworms of polycyclic aromatic hydrocarbons in soi l. Chemosphere. 48: 35-42.
ter Laak, T.L. ; Barendregt, A.; Hermens, lL.P. 2006. Freely dissolved pore water
concentrations and sorption coefficients of P AHs in spiked, aged, and field-
contaminated soils. Environ. Sei. Teehnol. 40: 2184-2190.
Thorsen, W.A.; Cope, W.G.; Shea, D. 2004. Bioavailability of PAHs: effects of soot
carbon and PAH source. Environ. Sei. Teehnol. 38: 2029-2037.
Tsai, P. ; Hoenicke, R. ; Yee, D.; Bamford, H.A.; Baker, lE. 2002 . Atmospheric
concentrations and fluxes of organic compounds in the northern San Francisco Estuary.
Environ. Sei. Teehnol. 36: 4741-4747.
164
van der Wal , L.; van Gestel, C.A.M.; Hermens, lL.M. 2004. Solid phase
microextraction as a tool to predict internaI concentrations of soil contaminants in
terrestrial organisms after exposure to a laboratory standard soil. Chemosphere. 54:
561-568.
Van Leeuwen, c.J. ; Hermens, lL.M. 1995. Ecotoxicological effects. Dans Risk
Assessment ofChemicals: An Introduction, Van Leeuwen, c.J. ; Hermens, lL.M. (Eds),
Kluwer Academic Publishers, Dordrecht, The Netherlands, 175-238.
Weissenfels, W.D.; Klewer, H.-l; Langhoff, l 1992. Adsorption of polycyclic aromatic
hydrocarbons (PAHs) by soi l particles: influence on biodegradability and biotoxicity.
Appl. Microbiol. Biotechnol. 36: 689-696.
Weston, D.P. ; Penry, D.L. ; Gulman, L.K. 2000. The role of ingestion as a route of
contaminant bioaccumulation in a deposit-feeding polychaete. Arch. Environ. Con/am.
Toxicol. 38: 446-454.
White, lC.; Alexander, M.; Pignatello, J.J. 1999a. Enhancing the bioavailability of
organic compounds sequestrated in soil and aquifer solids. Environ. Toxicol. Chem. 18:
182-187.
White, J.C.; Hunter, M. ; Pignatello, ll; Alexander, M. 1999b. Increase in
bioavailability of aged phenanthrene in soils by competitive displacement with pyrene.
Environ. Toxicol. Chem. 18: 1728-1732.
Wingo, C. W. 1966. Persistence and degradation of dieldrin and heptachlor in soil and
effects on plants. Agric. Exp. Sta. , Res. Bull. 914, 27 pp.
165
World Health Organization. 1998. Seleeted Nonheteroeyclie Polyeyclie Aromatie
Hydroearbons , Environrnental Health Criteria, No 202, 883 pp.
Yunker, M.B.; Macdonald, R.W.; Vingarzan, R.; Mitchell, R.H.; Goyette, D. ; Sylvestre,
S. 2002. PAHs in Fraser River Basin: a critical appraisal of PAH ratios as indicators of
PAH source and composition. Org. Geoehem. 32: 489-515.
Zhao, x.; Voice, T.C. 2000. Assessment ofbioavailability using a multicolumn system.
Environ. Sei. Teehnol. 34: 1506-1512.
Zimmerman, J.R.; Ghosh, u.; Millward, R.N.; Bridges, T.S.; Luthy, R.G. 2004.
Addition of carbon sorbents to reduce PCB and PAH bioavailability in marine
sediments: physicochemical tests. Environ. Sei. Teehnol. 38: 5458-5464.